A fillet weld is a roughly triangular bead of weld metal that joins two surfaces meeting at an angle, most commonly at 90 degrees. It’s the most widely used weld type in structural steel fabrication, largely because it requires little or no edge preparation before welding. If you’ve ever looked at a steel shelf bracket welded to a post, or a pipe welded to a flat plate, the triangular weld sitting in that inside corner is a fillet weld.
How a Fillet Weld Is Shaped
In cross section, a fillet weld forms a triangle between the two pieces of metal it connects. That triangle has several named parts, and understanding them matters because they determine the weld’s strength and how it gets measured.
The root is the innermost point where the two base metals meet, deep inside the joint. The toes are the two points where the weld face transitions into the base metal on either side. The face is the exposed outer surface of the weld, the part you can see and touch. The legs are the two straight sides of the triangle running along each base metal surface from root to toe. And the throat is the shortest distance from the root to the face, measured perpendicular to a line connecting the two toes.
The throat is the single most important dimension. It represents the thinnest cross section of the weld, meaning it’s where failure would start under load. For a standard fillet weld with equal legs, the throat thickness equals the leg length divided by roughly 1.414 (the square root of 2). So a fillet weld with 1/2-inch legs has a throat of about 0.354 inches. Engineers use throat thickness, not leg length, when calculating how much load the weld can carry.
Three Face Profiles
Not all fillet welds look the same from the outside. The face can be flat (called a mitre fillet), bulging outward (convex), or scooped inward (concave). All three are generally acceptable as long as the weld meets the required size. In structural welding, convexity doesn’t need to be measured and isn’t a pass/fail criterion on its own. Concavity is also acceptable, provided the throat thickness still meets the minimum specified on the drawing.
A convex profile adds extra metal beyond what’s structurally needed, which can sometimes create stress concentrations at the toes under cyclic loading. A concave profile looks cleaner and can actually perform better in fatigue, but it reduces the effective throat. In practice, a flat or slightly convex face is what most welders aim for.
Where Fillet Welds Are Used
Fillet welds work on any joint where two surfaces form an angle you can fill. The three most common configurations are:
- T-joints: One plate stands upright on another, forming a T shape. This is the classic fillet weld setup and appears constantly in structural steel, equipment frames, and machine bases. Pipes or tubes welded onto a base plate also form T-joints.
- Lap joints: Two plates overlap each other, and a fillet weld runs along the edge of the top plate where it sits on the bottom one. Common in sheet metal work and when extending or reinforcing plate structures.
- Corner joints: Two plates meet at their edges to form an L or a box shape. The inside corner is a natural home for a fillet weld. If more strength is needed, a second fillet can be added on the outside.
You can place fillet welds on one side of a joint or both. A single-sided fillet weld on a T-joint is vulnerable to bending forces that pry open the root. Adding a fillet on the opposite side eliminates that weakness, because the second weld prevents the first from being pulled in tension at the root.
Why Fillet Welds Are So Common
The main advantage over groove welds (where the base metal edges are beveled or shaped to create a channel that gets filled with weld metal) is simplicity. Fillet welds need no special edge preparation. You simply position the two pieces and weld the corner. That saves significant time and money in fabrication, especially on large structures with hundreds or thousands of joints.
Groove welds do carry higher loads per unit of weld because the weld metal fuses across the full thickness of the base metal. But that extra strength is often unnecessary. If a joint only needs to handle moderate loads, using a groove weld is, as welding professionals put it, “a waste of time.” Fillet welds sized correctly for the load are the more economical choice. When a joint needs more capacity than a fillet alone provides, engineers sometimes specify a groove weld on one side with a reinforcing fillet on the other, combining full penetration with additional cross-sectional area.
Sizing and Minimum Requirements
Fillet weld size is specified by leg length on engineering drawings. Common sizes range from 3/16 inch up to 3/4 inch or larger, depending on the thickness of the base metals and the loads involved. Structural welding codes, such as AWS D1.1, set minimum fillet weld sizes based on the thicker piece of base metal being joined. Thicker base metals require larger minimum welds because they pull heat away from the joint faster, and a weld that’s too small relative to the surrounding metal is prone to cracking as it cools.
The root pass (the first bead laid down) must be at least 1/8 inch to prevent cracking. From there, the welder builds up to the specified size, sometimes in multiple passes for larger welds. A 1/4-inch fillet can typically be done in a single pass, while anything above 5/16 inch usually requires two or more passes.
Reading Fillet Welds on Drawings
On blueprints, a fillet weld is indicated by a small right triangle placed on a horizontal reference line. The triangle’s position tells you which side of the joint gets the weld: below the reference line means the weld goes on the arrow side (the side the arrow points to), and above the line means it goes on the far side. Triangles on both sides mean fillet welds on both sides of the joint. A number next to the triangle indicates the leg size.
For more complex joints, like a beveled corner with reinforcing fillets, the drawing may combine groove and fillet symbols on the same reference line, or the engineer may include a detail sketch showing the exact joint geometry. When standard symbols can’t fully capture the intent, a note in the drawing’s tail referencing a separate sketch is common practice.
Common Defects to Watch For
The most frequent quality issue in fillet welds is undercut, a groove melted into the base metal at the toe of the weld that isn’t filled by weld metal. It reduces the thickness of the base material right at the stress concentration point where the weld meets the plate, weakening the joint. Undercut is usually caused by too much heat, too fast a travel speed, or improper torch angle.
Overlap (sometimes called cold lap) is the opposite problem: weld metal rolls over onto the base metal without actually fusing to it. It looks like the weld is bigger than it is, but the excess metal isn’t bonded and provides no strength. Incomplete fusion at the root is another concern, especially on single-pass welds. If the arc doesn’t penetrate deep enough into the corner, the weld’s effective throat is smaller than it appears, and the joint won’t carry its designed load.
Inspectors evaluate fillet welds primarily by visual examination, checking that the legs are the correct size, the profile is acceptable, the toes transition smoothly into the base metal, and there’s no visible cracking, porosity, or undercut. For critical applications, additional testing like ultrasonic or magnetic particle inspection may follow, but the visual check catches the majority of problems.