Fillet Height: How to Measure and Specify It

Fillet height is the distance the solder rises up the side of a component termination, measured from the surface of the pad to the top of the solder at the point where it meets the part. It is the dimension that visual inspection approximates and that a cross section measures directly. Fillet height tells you how much metal is supporting the joint, and therefore how much of the thermal and mechanical load it can carry.

This guide covers what fillet height describes, how to measure it with and without sectioning, how the wetting angle relates to it, and how to write acceptance criteria an inspector can apply. It applies to gull-wing leads, chip components and through-hole joints, with different limits for each.

<img src="https://www.gopcba.com/wp-content/uploads/2026/06/harsh-environment.jpg" alt="Fillet height measured on a cross section through a solder joint” />

What Fillet Height Describes

A solder joint consists of the interface at the pad, the interface at the termination, and the fillet that connects them. Fillet height describes how far that connecting metal climbs the termination. A high fillet wraps more of the termination, so the load is spread over a larger area and the joint resists peeling better.

Fillet height is not the same as solder volume. A joint can have plenty of solder and still show a low fillet if the metal has spread along the pad rather than rising up the lead. That is why the two are specified separately, and why a joint that looks generously filled can still fail on fillet height.

Why Fillet Height Matters Mechanically

The load on a surface mount joint is largely a peel and shear combination produced by thermal expansion mismatch. The peel load concentrates at the toe of the fillet, which is the point furthest from the pad. A taller fillet moves the toe further up the termination and increases the length over which the stress is distributed.

A low fillet concentrates the same load into a shorter region, so the stress at the toe is higher. That is why a joint with a low fillet is more likely to crack under thermal cycling even when it is electrically sound and passes a pull test at room temperature. Our notes on thermal cycling describe the failure pattern.

Measuring Fillet Height by Cross Section

The cross section is the reference method. A sample is potted, ground and polished through the centre of the joint, and the fillet height is measured on the image against the pad surface. The measurement should be taken at the toe and at the heel, because the two differ and the lower value controls the acceptance.

Preparation matters. Grinding that smears the soft solder obscures the boundary, so the final steps should be light and the etching chosen for the alloy. A measured value taken from a smeared section can be wrong by a large fraction, and the error is not visible unless a reference joint is prepared alongside.

Measuring Without Sectioning

Optical measurement uses a low-angle view and a calibrated reference, or a laser profilometer that maps the joint surface. Both give a height relative to the pad, and both need the pad surface to be visible at the same time as the fillet. Where the joint is behind a component, neither method works.

The optical method is quicker and non-destructive, and it is accurate enough to screen a joint that is clearly short. It is less accurate at the toe, where the fillet meets the pad at a shallow angle, because the transition is gradual. Where the limit is tight, the optical result should be confirmed by a cross section on a sample.

Wetting Angle and Its Relationship to Height

The wetting angle is the angle between the solder surface and the pad at the point where the two meet. A low wetting angle means the solder has feathered onto the pad, and it is generally associated with a fillet that has spread rather than risen. A high wetting angle means the solder has not spread, which can produce a tall but narrow fillet.

The two measurements therefore describe different things and should not be substituted for each other. A joint can pass a wetting test and fail a fillet height limit if the solder has spread along a long pad. A joint can show an acceptable height and a poor wetting angle if the solder has balled up on the termination. Our inspection standards notes cover both criteria.

Fillet Height on Different Component Types

A chip component has two terminations and the fillet rises on the end face of each. The limit is usually expressed as a fraction of the termination height, with a minimum that ensures the joint wraps enough of the part. A gull-wing lead has a fillet at the toe and at the heel, and the heel fillet is the one that carries the peel load.

A through-hole joint has a fillet on both sides of the board, and the requirement is usually expressed as a percentage of barrel fill together with a visible fillet on each side. A ball grid array has no external fillet at all, so the criteria shift to the collapse of the ball and to the void fraction. This is why the limit belongs to the joint family rather than to the product.

Writing Acceptance Criteria

The criterion should state the measurement method, the reference surface and the minimum value. A typical form for a chip component is that the fillet must reach at least 25 % of the termination height, measured by cross section, at both ends. For a gull-wing lead, a common form is a visible concave fillet at the toe with a defined minimum height at the heel.

The criterion should also state what is done when the limit is not met. A joint below the minimum is a candidate for rework, and one that is marginal should be assessed against the load the joint carries rather than against a general rule. Recording the measured value rather than a pass or fail allows a trend to be seen across a lot, and it is the data that shows whether the process is drifting.

Common Measurement Mistakes

The most common mistake is measuring at the wrong point. The fillet is highest where the termination meets the pad and lowest at the toe, and a measurement taken at the highest point overstates the result. The measurement should be taken at the location the criterion names, and the location should be marked on the image.

The second mistake is a reference surface that is not the pad. Measuring from the solder mask rather than from the copper adds the mask thickness to the result, which on a tall mask layer can be a significant fraction of the limit. Where the pad and the mask are at different levels, the reference has to be stated in the procedure rather than assumed.

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

FAQ

Is fillet height controlled by the stencil? It is controlled by the ratio of solder volume to pad area and by the pad extension. A thicker deposit raises the fillet only if the pad is narrow enough to force the solder upwards, which is why the land pattern matters as much as the stencil.

Can fillet height be measured on a production unit? It can be estimated optically without damage. A true measurement needs a section, so it is applied to samples and to the first article rather than to every board.

What if the fillet is tall but the wetting angle is poor? That combination indicates a joint that has not bonded over the full pad, so the apparent height is misleading. The joint should be assessed on the interface rather than on the fillet, and a cross section is the way to see it.