Minimum Weld Size Reference
Reference minimum fillet weld sizing concepts by thickness. Confirm with WPS and governing standards.
Why minimum weld sizes exist
Minimum fillet weld sizes are not arbitrary -- they exist to prevent cracking caused by rapid cooling. When a small weld is deposited on a thick base metal, the large mass of surrounding steel acts as a heat sink, pulling heat away from the weld zone very quickly. This rapid cooling produces a hard, brittle heat-affected zone (HAZ) prone to hydrogen-induced cracking. By mandating a minimum weld leg size that increases with the thickness of the thicker part joined, steel design codes ensure that enough heat is delivered to slow the cooling rate and reduce cracking risk.
The governing thickness for minimum weld size is the thicker of the two parts being joined, not the thinner one. This is a frequent point of confusion. If you are welding a 10 mm gusset plate to a 25 mm thick column flange, the minimum fillet weld size is controlled by the 25 mm flange, not the 10 mm gusset. The logic is straightforward: the thicker part has more thermal mass and drives the cooling rate.
Different codes express these minimums in slightly different formats but follow the same metallurgical principle. AISC 360 Table J2.4 specifies minimum fillet weld sizes based on the thickness of the thinner part joined (note: AISC uses the thinner part as the index, but the practical effect is similar). AS 4100 Table 9.7.3.2 indexes against the thickness of the thicker part. EN 1993-1-8 addresses the issue indirectly through throat thickness requirements and references to EN 1011 for welding procedure controls. In all cases, the designer must also comply with the Welding Procedure Specification (WPS), which may impose additional constraints based on preheat, interpass temperature, and consumable selection.
Key concepts and common mistakes
When specifying or checking fillet weld sizes, keep these points in mind:
- Minimum size is a lower bound, not a design target. The required weld size from a strength calculation will often exceed the minimum. The minimum only matters when the strength-required size is very small.
- Which thickness governs depends on your code. AISC 360 indexes the minimum against the thinner part; AS 4100 indexes against the thicker part. Check which convention your code uses before looking up the table.
- Maximum weld size on an edge. Codes also limit the maximum fillet weld size relative to the thickness of the part at a free edge -- typically the material thickness minus 1/16" (2 mm) for parts 1/4" (6 mm) or thicker. This prevents the weld toe from melting past the edge.
- Single-pass vs. multi-pass. Minimum size requirements assume the weld can be deposited in a single pass or that a multi-pass procedure provides equivalent heat input. A very small weld on thick material may not physically deliver enough heat, even if the leg size technically meets the table.
- WPS governs in practice. The Welding Procedure Specification for the project will specify preheat temperatures, interpass temperatures, and heat input ranges. These controls achieve the same goal as the minimum weld size table -- preventing rapid cooling -- and may override the table minimum in either direction.
- Do not confuse leg size with throat size. The minimum weld size tables specify the leg dimension (w). The effective throat for strength calculations is 0.707w for equal-leg fillets. Mixing up leg and throat will produce a 30% error in calculated capacity.
For the full verification and documentation workflow, see How to verify calculator results.
FAQ
Why can I not just use the smallest possible weld everywhere? A weld that is too small relative to the base metal thickness will cool too rapidly, creating a brittle heat-affected zone susceptible to cracking. The minimum size tables in design codes are calibrated to ensure sufficient heat input for sound metallurgy. Undersized welds may pass a visual inspection but fail in service under fatigue or impact loading.
Which part thickness governs the minimum -- the thicker or the thinner? It depends on your code. AISC 360 Table J2.4 uses the thickness of the thinner part joined as the index column. AS 4100 Table 9.7.3.2 uses the thicker part. Both approaches achieve the same objective (preventing cold cracking), but you must use the convention specified in your governing standard.
How do minimum weld sizes differ between AISC, AS 4100, and EN 1993? AISC 360 and AS 4100 both provide explicit tables with thickness brackets and corresponding minimum leg sizes, but they index against different parts (thinner vs. thicker). EN 1993-1-8 does not include a direct equivalent table; instead, it relies on minimum throat thickness requirements and references to EN 1011 welding procedure standards for heat input control. The practical minimums are similar across codes, but always look up the specific table for your standard.
Is there also a maximum fillet weld size? Yes. Along a free edge of material 1/4" (6 mm) or thicker, the maximum fillet weld leg size is the material thickness minus 1/16" (2 mm). This prevents the weld from overrunning the edge. Along edges thinner than 1/4", the weld may be the full thickness of the edge. For welds not at an edge, there is no geometric maximum, but practical and economic limits apply.
How does the WPS relate to minimum weld size tables? The WPS controls the actual welding parameters -- voltage, current, travel speed, preheat, interpass temperature -- that determine heat input. If the WPS provides adequate preheat for a thick joint, the cooling rate may be acceptable even with a weld smaller than the code table minimum. In practice, designers specify the code minimum as a floor, and the WPS provides the procedural controls to ensure sound welds at or above that floor.
Related pages
- Welded connections calculator
- Weld symbol generator
- Gusset plate calculator
- Steel grades reference
- Unit converter
- Tools directory
- Reference tables directory
- Guides and checklists
- How to verify calculator results
- Disclaimer (educational use only)
Disclaimer (educational use only)
This page is provided for general technical information and educational use only. It does not constitute professional engineering advice, a design service, or a substitute for an independent review by a qualified structural engineer. Any calculations, outputs, examples, and workflows discussed here are simplified descriptions intended to support understanding and preliminary estimation.
All real-world structural design depends on project-specific factors (loads, combinations, stability, detailing, fabrication, erection, tolerances, site conditions, and the governing standard and project specification). You are responsible for verifying inputs, validating results with an independent method, checking constructability and code compliance, and obtaining professional sign-off where required.
The site operator provides the content "as is" and "as available" without warranties of any kind. To the maximum extent permitted by law, the operator disclaims liability for any loss or damage arising from the use of, or reliance on, this page or any linked tools.