Types of Welding in Steel Construction Explained

Steel buildings are held together by bolts and welds. Bolts dominate on site, but most of the welding is done earlier, in the fabrication shop, where plates, stiffeners and end plates are joined to beams and columns before they are delivered. Understanding the main types of welding helps explain why fabricators choose particular processes, why some welds are done in the shop and others in the field, and why weld quality is inspected so carefully.

This article explains what happens in a weld, the processes most common in structural steelwork, the difference between fillet and butt welds, a simple example of how a weld is sized, how welds are inspected, and the mistakes that cause problems. It is written for students and for building owners who want to read a steel fabrication specification with some confidence.

What a weld actually is

Fusion welding melts the edges of the parts being joined, together with a filler metal, so that they flow into a common pool and solidify as one continuous piece of steel. In structural work the heat comes from an electric arc struck between an electrode and the workpiece, reaching temperatures high enough to melt steel almost instantly. The molten pool must be protected from the oxygen and nitrogen in air, which would otherwise make the weld porous and brittle, so every arc process uses some form of shielding: a flux coating, a granular flux, or a shielding gas.

A properly made weld is as strong as, or stronger than, the parent steel. But the process also introduces heat, distortion, residual stress and a heat-affected zone in the surrounding steel whose properties are altered. Much of welding engineering is about controlling these side effects through the right procedure, consumables, preheat and welder skill.

The main types of welding in steel construction

The international standard EN ISO 4063 gives each welding process a reference number, which is how processes are usually identified on welding procedure documents. The ones most often met in structural steelwork are as follows.

Manual metal arc welding (stick welding, process 111)

Also called shielded metal arc welding, this uses a consumable electrode rod coated in flux. The flux burns to produce a shielding gas and a slag that covers the weld. Equipment is simple and portable and works outdoors in wind, so it remains common for site welding and repairs. It is slower than wire processes because the welder must stop to change rods and chip off slag.

Metal active gas welding (MAG or MIG, process 135)

A continuously fed wire electrode is shielded by a gas, typically argon mixed with carbon dioxide for steel. It is fast, clean and well suited to shop fabrication, where draughts that would blow the gas shield away can be controlled. In everyday speech it is usually called MIG welding, although for steel the correct term is MAG.

Flux-cored arc welding (process 136; 114 when self-shielded)

Similar to MAG, but the wire is tubular and filled with flux. Some variants also use shielding gas; self-shielded versions do not, which makes them useful on site. It gives high deposition rates and good penetration and is widely used for heavier structural sections.

Submerged arc welding (process 121)

The arc burns beneath a blanket of granular flux, so it is invisible and produces no spatter. It is usually mechanised, running along long straight seams, and deposits large volumes of weld metal at high quality. Fabricators use it for plate girders, built-up columns and the long web-to-flange welds of fabricated sections.

Tungsten inert gas welding (TIG, process 141)

A non-consumable tungsten electrode creates the arc while filler is added separately. TIG produces very precise, clean welds but is slow, so in structural work it is mainly used for thin material, stainless steel, root runs and architectural details rather than primary members.

Stud welding

Arc stud welding fixes headed shear studs to steel beams, often through the metal deck, in a fraction of a second using a special gun. These studs are what make steel beams and concrete slabs act together in composite construction and are tested on site by bending sample studs. Our article on metal decking explains how they fit into a composite floor.

Weld joint types: fillet and butt welds

Separate from the welding process is the form of the weld itself. The two main forms in structural steelwork are:

Fillet welds — a roughly triangular weld laid in the corner between two plates meeting at an angle, typically a T-joint or lap joint. They need little edge preparation and are the most common weld in steel buildings, used for end plates, fin plates, stiffeners and bracing gussets.
Butt welds — made between the prepared edges of two plates, usually in the same plane, so the weld fills the joint through its thickness. A full-penetration butt weld develops the full strength of the plate and is used where the joint must carry the member’s full capacity, such as flange splices or moment connections.

Partial-penetration butt welds, plug welds and slot welds also exist for particular situations. The choice between fillet and butt welds, and between shop and site welding, strongly influences connection design. Our sibling article on bolted versus welded connections compares the approaches, and our guide to the moment connection shows where full-strength welds are typically needed.

A simple example: sizing a fillet weld

A fillet weld’s strength depends on its throat thickness, a, the shortest distance from the root to the face of the weld. For an equal-leg fillet weld, the throat is approximately 0.7 times the leg length. So a 6 mm leg fillet has a throat of about 0.7 × 6 ≈ 4.2 mm.

Under the simplified method in Eurocode 3, the design resistance per unit length is the throat thickness multiplied by a design shear strength of the weld, which depends on the steel’s ultimate tensile strength and code factors. Taking an illustrative design strength of about 250 N/mm² for welds on S355 steel, a 4.2 mm throat resists about 4.2 × 250 ≈ 1,050 N per millimetre of weld, or roughly 1 kN/mm. Two lines of weld, each 150 mm long, would therefore resist in the order of 300 kN.

The example shows two points. First, weld capacity scales directly with throat and length, so a slightly undersized weld loses capacity in proportion. Second, a bigger weld is not always better: oversized welds add heat, distortion and cost without benefit. Actual design values must come from the code and the steel grade, discussed in our guide to steel grades in construction.

Quality, qualification and inspection

Because a weld’s interior cannot be seen, quality relies on control before and during welding as well as inspection afterwards. Welding procedures are qualified by testing, commonly under EN ISO 15614-1, and welders are qualified to carry out those procedures, commonly under EN ISO 9606-1. On Eurocode projects the execution of steelwork, including welding, is typically specified to EN 1090-2, while projects to American practice reference the AWS D1.1 Structural Welding Code.

Inspection begins with visual examination of every weld for size, profile, cracks and undercut. Selected welds then receive non-destructive testing: magnetic particle or dye penetrant testing to find surface cracks, and ultrasonic or radiographic testing to find internal defects in butt welds. Acceptance criteria are commonly set by reference to the quality levels of EN ISO 5817.

Common mistakes

Site welding when shop welding is possible. Site welds are harder to position, shield and inspect. Designs usually aim for shop-welded, site-bolted connections.
Welding galvanised or painted steel without preparation. Coatings contaminate the weld and produce hazardous fumes; they must be removed and the area reinstated afterwards.
Ignoring preheat on thick or high-strength steel. Cooling too fast can cause hydrogen cracking in the heat-affected zone, sometimes hours after welding.
Unspecified weld sizes. Drawings should state type, size and length, not just “weld all round”.
Welding to existing structures without assessment. Older steels may have unknown chemistry and poor weldability, so testing is often required first.

Frequently asked questions

What are the main types of welding used in construction?

The most common in structural steelwork are manual metal arc (stick) welding, metal active gas (MAG or MIG) welding, flux-cored arc welding and submerged arc welding, with TIG used for thin or precise work and arc stud welding used to fix shear studs for composite floors.

Which welding type is strongest?

No process is inherently strongest. A correctly made weld from any qualified procedure can match the parent steel. Strength depends on the weld type and size, the consumables, the procedure and the welder’s skill rather than the process name.

What is the difference between a fillet weld and a butt weld?

A fillet weld is a triangular weld laid in the corner between two plates meeting at an angle, while a butt weld joins the prepared edges of plates so the weld fills the joint through its thickness. Full-penetration butt welds can develop the full strength of the plate.

Why is MIG welding used more in fabrication shops than on site?

MIG or MAG welding relies on a gas shield that wind can easily disrupt, causing porosity. In a shop the environment is controlled. On site, stick welding or self-shielded flux-cored welding is often preferred because the shielding comes from the flux.

How are welds inspected?

All welds are visually inspected for size, shape and surface defects. Critical welds are also tested non-destructively, using magnetic particle or dye penetrant methods for surface cracks and ultrasonic or radiographic methods for internal flaws, against defined acceptance criteria.

Need a steel connection, welded alteration or fabrication specification reviewed? Talk to our consultancy or get in touch.