How to Measure Weld Size: Fillet and Groove Welds

Weld size is measured differently depending on whether you’re looking at a fillet weld or a groove (butt) weld, but in both cases, the key dimensions are leg length, throat thickness, and face width. For the most common type, a fillet weld on a 90-degree joint, the theoretical throat equals 0.707 times the leg length. Understanding that relationship is the foundation of every weld size measurement you’ll make.

Key Dimensions of a Fillet Weld

A fillet weld sits in the corner where two pieces of metal meet, and its cross-section forms a roughly triangular shape. Three measurements define its size:

  • Leg length: The distance from the weld root (the inside corner where the two pieces meet) to the weld toe (where the weld surface meets the base metal) on each side. A fillet weld has two legs, and they aren’t always the same length.
  • Throat thickness: The shortest distance from the root to the face of the weld. This is the dimension that actually determines the weld’s load-carrying capacity.
  • Face width: The distance across the exposed surface of the weld. On an equal-leg fillet in a 90-degree joint, the face width is roughly twice the throat thickness.

On most prints and welding symbols, the specified size refers to leg length. But the throat is what engineers care about for strength calculations. For an equal-leg fillet weld on a 90-degree joint, you convert between them by multiplying the leg length by 0.707. So a 1/4-inch fillet weld (leg size) has a theoretical throat of about 0.177 inches. This conversion only works cleanly when the joint angle is 90 degrees and the legs are equal. Unequal legs or non-perpendicular joints require trigonometry.

How Weld Shape Affects Measurement

Not all fillet welds have a flat face. The profile can be convex (bulging outward), concave (curving inward), or roughly flat, and the shape changes how you measure.

For a convex weld, you measure each leg individually: the distance from the root to the toe on both sides. The extra metal bulging past the theoretical triangle doesn’t count toward the effective throat in most codes, so leg length is your primary measurement. Both legs need to meet the minimum specified size, because if one leg is short, the throat will be undersized even though the weld looks beefy from the convex face.

For a concave weld, the face curves inward, which means the actual throat is shorter than what the leg dimensions alone would suggest. In this case, you measure the throat directly rather than relying on leg dimensions. Fillet weld gauges with a curved (double-arc) side are designed specifically for this. You place the gauge against the weld face and read the throat dimension, ignoring the apparent leg sizes.

Measuring only one dimension without considering the weld shape is unreliable. A weld with generous legs but a deeply concave face can have an inadequate throat. A convex weld with one short leg can look fine from the face but fail to meet spec.

Tools for Measuring Weld Size

The tools welders and inspectors use range from simple stamped-metal gauges to multipurpose instruments. Each has strengths and trade-offs.

Fillet weld gauge sets are the simplest option. These are small, flat pieces of metal cut to specific sizes (typically in 1/16-inch increments) that you hold against the weld to check leg length or throat. You find the gauge that fits flush. They’re inexpensive, durable, and hard to misuse. For routine visual inspection of fillet welds, a standard set of fillet gauges handles most situations.

Bridge Cam gauges (sometimes called Cambridge gauges) are the Swiss Army knife of weld inspection. A single tool can measure fillet leg size, throat, weld reinforcement height on butt joints, undercut depth, and compare groove angles. The gauge uses sliding parts and thumb screws to take readings. Experienced inspectors often reach for the Bridge Cam first because it handles so many tasks. The downside is that the small fasteners can loosen over time, and the tool requires more care to get accurate, repeatable readings than a simple fillet gauge.

V-WAC undercut gauges measure the depth of undercut, the groove sometimes left along the weld toe where the base metal has been melted away. Undercut doesn’t directly measure weld size, but it affects whether a weld passes inspection.

For most visual weld inspections, a practical kit includes a set of fillet gauges, a V-WAC gauge, a pair of dividers, and a 6-inch steel ruler. Multipurpose gauges like the Bridge Cam are useful additions but work best as supplements rather than replacements for dedicated tools.

Measuring Groove (Butt) Welds

Groove welds join two pieces edge-to-edge, and their size is defined by penetration depth and reinforcement height. Full-penetration groove welds should extend through the entire thickness of the base metal. Partial-penetration groove welds have a specified depth.

Reinforcement is the extra weld metal that sits above the surface of the base metal. You can measure it with a Bridge Cam gauge or a straight edge and ruler, placing the straight edge across the base metal on both sides and measuring the height of the weld crown. For this measurement to be accurate, the base metal surfaces need to be reasonably flat and aligned.

Penetration depth, the dimension below the surface, can’t be measured visually. If the joint is welded from one side only and you can see the root side, you can measure root penetration (how far the weld protrudes through the back). But for confirming full penetration inside a closed joint, you need non-destructive testing.

Checking Internal Dimensions With Ultrasonics

Visual gauges only tell you what’s happening on the surface. For internal weld size, particularly penetration depth, ultrasonic testing (UT) is the standard non-destructive method. UT sends high-frequency sound waves into the weld and reads the reflections to map the weld’s internal shape, detect lack of fusion, and estimate penetration.

Research has shown that ultrasonic energy measurements correlate well with weld shape, and the technique has been used for decades to characterize penetration depth and locate internal flaws. A frequency of at least 1 MHz is typically needed for meaningful boundary detection between weld metal and base metal. Lower frequencies (around 500 kHz) lack the resolution to distinguish the boundary reliably.

UT is particularly important for deep-penetration fillet welds, where extra root penetration adds to the effective throat but is completely invisible from the outside. You simply cannot measure the throat of a deep-penetration weld with a surface gauge.

Size Tolerances in Common Codes

Weld size isn’t pass/fail at a single number. Codes build in tolerances and minimum requirements based on material thickness and loading conditions.

Under AWS D1.1, the structural steel welding code widely used in North America, fillet welds can vary from convex to concave in profile. If the gap between the parts being joined exceeds 1/16 inch, each leg of the fillet weld must be increased by the amount of that gap. The maximum allowable gap is 3/16 inch. Undercut (the groove melted into the base metal at the weld toe) is limited to 1/32 inch for most joints, tightening to 0.01 inch for primary members under cyclic tension loading. Minimum fillet weld sizes are set according to the thickness of the thicker piece being joined, and intermittent fillet welds must be at least 1.5 inches long.

ISO 5817 takes a different approach, defining three quality levels (B, C, and D) with progressively tighter limits. Level D is the most permissive, Level B the strictest. For excess weld metal on a butt weld, Level B caps reinforcement height at 0.1 times the weld width plus 1 mm, with a maximum of 5 mm. Level D allows 0.25 times the width plus 1 mm, up to 10 mm. Linear misalignment between plates follows a similar pattern: Level B allows misalignment up to 0.1 times the plate thickness (max 3 mm), while Level D allows 0.25 times thickness (max 5 mm).

Your drawings and welding procedure specifications will reference which code applies. The tolerances determine not just whether a weld is the right size, but how much variation is acceptable before rework is required.

Step-by-Step Fillet Weld Measurement

To measure a fillet weld on a standard 90-degree T-joint, start by checking both legs. Place a fillet gauge or the leg-measuring edge of a Bridge Cam flat against one base metal surface, with the gauge tip touching the weld toe on the opposite side. Read the leg length. Repeat on the other leg. Both legs should meet or exceed the specified size.

Next, assess the profile. If the face is flat or convex, the leg measurements give you what you need. Calculate the theoretical throat by multiplying the smaller leg by 0.707. If the face is concave, measure the throat directly using the curved side of a fillet gauge or the throat-measuring function on your Bridge Cam. Compare that throat to the minimum required (0.707 times the specified leg size).

Check for undercut along both toes using a V-WAC gauge or the undercut scale on a Bridge Cam. Look at the weld length if it’s an intermittent weld. Finally, check the fit-up gap. If you can see daylight between the parts at the root exceeding 1/16 inch, the specified weld size should have been increased accordingly.