Lap joints are among the most common weld connections in structural steel fabrication—cover plates on plate girder flanges, reinforcing pads on erection columns, splice plates on built-up members, and base-plate doubler plates all use lapped configurations. Because lap joints are almost always made with fillet welds and are covered under the prequalified provisions of AWS D1.1:2025, shops sometimes treat them as low-documentation connections that need minimal WPS attention. That assumption is a liability.
AWS D1.1 has specific requirements for lap-joint geometry, minimum overlap, fillet weld size, end returns, and inspection. A WPS that fails to address these variables completely may pass a desk review but fail a third-party audit or owner inspection.
What Makes a Lap Joint Different from a T-Joint
In a T-joint the parts are roughly perpendicular—the web of a wide-flange to a flange plate, for example. In a lap joint the parts are parallel and overlapping, with one plate sitting directly on top of the other. Both joint types typically use fillet welds, and both can be prequalified under Clause 5, but the stress flow through a lap joint creates concerns that do not apply to a simple T-joint fillet:
- Eccentricity. In a double-lap joint loaded in tension, the load is transferred from one plate, through the fillet welds, into the overlapping plate. The centroidal axes of the two parts are offset, creating a secondary bending moment at the connection. This eccentricity is acceptable in structural design but means the weld root is subjected to peeling forces in addition to shear.
- End-termination stress concentration. The weld terminates at the edge of the lap, and stress concentrations at both ends of the connection can accelerate fatigue crack initiation in cyclically loaded applications. AWS D1.1 addresses this with end return requirements.
- Fit-up sensitivity. A lap joint with a gap between the plates changes the effective throat of the fillet weld and can affect the available weld size. The code's fit-up tolerances for lap joints are distinct from those for T-joints.
Understanding these differences is what allows a CWI or WPS preparer to write and inspect a lap-joint weld correctly.
Minimum Overlap Requirement
AWS D1.1:2025 requires that the overlap between the two parts in a lap joint be at least five times the thickness of the thinner member and not less than 1 in (25 mm). This requirement exists so that the connection has enough length to develop the required load transfer before the weld terminates.
A common shop mistake is fabricating lap-joint reinforcement plates that are sized only large enough for the required weld length without checking whether the overlap dimension satisfies this minimum. If the analysis calls for a 3-in long fillet weld on a 3/8-in plate, the minimum overlap is five times 3/8 in = 1-7/8 in; the 3-in weld length clears that requirement. But if the same weld is on a 1-in plate, the minimum overlap jumps to 5 in, and a plate cut to 4 in (thinking the weld length was sufficient) is non-conforming.
The CWI should verify the overlap dimension from the fabrication drawing before welding and confirm it matches what Table 7.7 and Clause 7 require for the actual plate thickness in service.
Rule library based on AWS D1.1:2025; verify against your governing edition.
Minimum Fillet Weld Size
AWS D1.1:2025 Table 7.7 sets the minimum fillet weld leg size for structural welds. For lap joints the minimum size is controlled by the thinner of the two parts joined when low-hydrogen processes are used. When non-low-hydrogen processes are specified on the WPS, the minimum size is governed by the thicker part—which can be more restrictive.
This distinction matters for WPS writing. A shop that uses SMAW with E6010 root passes on lap joints (a practice sometimes used to access tight-fit applications) should recognize that E6010 is not a low-hydrogen electrode, so the thicker-part column of Table 7.7 governs. If the WPS specifies a minimum fillet size that was calculated from the thinner-plate column, the minimum weld size stated on the document is under-stated for the specified process.
Minimum fillet weld size is a floor, not a target. The engineered design size—based on the connection's required shear strength—will often be larger, particularly on heavily loaded lap splices. The WPS should state both the minimum allowable size per Table 7.7 and the range qualified under the WPS; the welder applies the design size from the drawing.
For more detail on how Table 7.7 applies to typical structural connections, see minimum fillet weld size under AWS D1.1 Table 7.7.
End Returns for Tension-Loaded Connections
When a lap joint is loaded in axial tension—a tension strap, a splice plate on a tension chord, a cover plate—AWS D1.1:2025 Clause 7 requires that the fillet welds be returned around the end of the overlapping member. The end return welds wrap the corner of the plate and prevent the weld termination from peeling under load.
The required return length is at least twice the nominal weld size. For a 5/16-in fillet weld, the return must extend at least 5/8 in around the corner. This seems minor but is frequently skipped in shop fabrication, especially when the connection is built from a drawing that shows only the main weld length without explicitly calling out the returns.
WPS documents for tension-loaded lap joints should explicitly state whether end returns are required and what their minimum length is. A prequalified WPS that does not mention end returns leaves the welder without guidance when the AHJ or inspector asks for them.
End returns are not required for lap joints under compression-only or shear-dominant loading, but the engineer's connection design should confirm the loading condition. A CWI who is unsure whether end returns are required should check the connection's loading condition with the EOR before releasing the work.