An inclusion in welding is a solid foreign material trapped inside the weld metal during solidification. Unlike porosity, which involves gas pockets, inclusions are physical contaminants, most commonly pieces of slag, bits of tungsten electrode, or oxide films that failed to float to the surface before the weld cooled. They weaken the joint by acting as stress concentrators, and they’re one of the most common defects inspectors look for.
Types of Welding Inclusions
Inclusions fall into three main categories based on what gets trapped in the weld.
Slag inclusions are the most frequent type. They appear as small, elongated particles of non-metallic material embedded in the weld metal. They tend to be lumpy and sit near the weld surface, though they can occur deeper in multi-pass welds. Most are nearly invisible to the naked eye and require magnification to identify. Slag inclusions are especially common in stick welding (SMAW) and flux-cored arc welding (FCAW), where a flux coating produces a protective slag layer that must be removed between passes.
Tungsten inclusions are specific to TIG welding (GTAW), where a tungsten electrode is used to create the arc. The electrode is designed to be non-consumable, meaning it shouldn’t melt into the weld. But when the welder accidentally dips the electrode into the molten puddle, or when shielding gas coverage is incomplete, small pieces of tungsten break off and become embedded in the weld metal. Tungsten is extremely dense, so even tiny fragments show up clearly on inspection film.
Oxide inclusions form when metal oxides on the surface of the base material or filler wire get folded into the weld pool instead of being cleaned away. This is a particular problem with aluminum, which forms a stubborn oxide layer almost instantly when exposed to air. That oxide film melts at a much higher temperature than the aluminum underneath, so it doesn’t simply dissolve during welding. It has to be mechanically or chemically removed beforehand. Shielding gases with higher oxygen or carbon dioxide content also increase oxide inclusions by reacting with elements like manganese and silicon in the weld, reducing both the strength and toughness of the finished joint.
What Causes Inclusions
The root causes vary by inclusion type, but most come down to inadequate cleaning, poor technique, or incorrect settings.
For slag inclusions, the primary trigger is failing to remove slag between weld passes. When two adjacent weld beads are deposited without enough overlap, a void forms between them. Slag fills that void, and when the next layer is deposited on top, the trapped slag doesn’t melt out. Undercut along the sidewall, caused by too much current paired with high travel speed, creates crevices that are especially difficult to clean. Uneven joint gaps compound the problem by creating pockets where slag can settle.
Tungsten inclusions result from electrode contact with the weld puddle. Apprentice welders are taught to maintain a gap of 4 to 6 millimeters between the electrode tip and the workpiece, but in practice, momentary contact happens. Research on apprentice welders confirmed that the composition of welding fume particles shifted noticeably when welders repeatedly touched the electrode to the puddle, depositing tungsten and cerium from the electrode tip into the weld.
Oxide inclusions trace back to surface contamination. Dirt, mill scale, and oxide films left on the base metal or filler wire introduce non-metallic material into the weld pool. With aluminum, even freshly cleaned surfaces re-oxidize within seconds, so timing matters as much as technique.
How Inclusions Affect the Weld
Inclusions act as internal stress concentrators. Under repeated loading, a tiny inclusion can become the starting point for a fatigue crack. Welded joints already exhibit greater variability in fatigue life than base metal because of the combined effects of residual stress, microstructural changes from heat, and geometric stress concentrations at the weld toe. Adding inclusions to that mix increases uncertainty further. Two identical-looking welds tested at the same stress level can have significantly different fatigue lives if one contains inclusions and the other doesn’t.
In structural and safety-critical applications like bridges, pressure vessels, and pipelines, inclusions above a certain size or frequency are grounds for rejection and mandatory repair. Even in less critical work, inclusions reduce the effective cross-section of the weld and can compromise corrosion resistance if they sit near the surface.
How Inclusions Are Detected
Because most inclusions are too small to see on the weld surface, inspectors rely on non-destructive testing to find them.
Radiographic testing (X-ray) passes radiation through the weld and captures an image on film or a digital detector. Low-density inclusions like slag appear as dark spots on the film, while high-density inclusions like tungsten appear as bright spots. It’s a reliable method for detecting inclusions, porosity, cracks, and voids in the interior of a weld, though it’s relatively slow and expensive.
Ultrasonic testing sends high-frequency sound waves through the weld and listens for reflections. Because sound travels at a nearly constant speed through a given material, the timing of reflected signals reveals the exact position and depth of a discontinuity. Ultrasonic testing is faster than radiography for many applications and doesn’t involve radiation, making it practical for field inspections.
Preventing Inclusions
Most inclusion prevention comes down to disciplined cleaning and consistent technique.
- Clean between every pass. Grinding is the most thorough method, especially in narrow butt joints or when undercutting has created slag traps. For simpler joints, wire brushing or light chipping can be sufficient.
- Maintain even joint gaps. Uneven spacing between workpieces creates cavities that collect slag. Consistent fit-up produces consistent results.
- Plan multi-pass welds carefully. Line up your weld beads to minimize gaps between them. Each bead should overlap the previous one enough to avoid forming pockets.
- Keep a consistent arc length. For stick welding, the distance between the electrode tip and the base metal should roughly equal the core diameter of the electrode. A 1/8-inch electrode calls for a 1/8-inch arc length. This delivers steady voltage that pushes flux to the weld surface where it belongs.
- Check your electrode. If slag builds up on the tip of a stick electrode, file it off before continuing. Welding with a contaminated electrode introduces inclusions directly.
- Avoid electrode contact in TIG welding. Keep the torch 4 to 6 millimeters above the workpiece. If you do dip the electrode, stop, regrind the tip to a clean point, and resume.
- Remove oxide layers before welding aluminum. Mechanical cleaning (stainless steel brush or abrasive pad) immediately before welding gives the best results, since the oxide reforms quickly.
- Use appropriate shielding gas. For non-ferrous metals like aluminum, pure argon or argon-helium blends minimize oxide formation in the weld pool. Shielding gases containing oxygen or carbon dioxide increase oxide inclusions.
How Inclusions Are Repaired
When inclusions are found during inspection, the standard repair process involves excavating the defective area, verifying the defect is fully removed, and re-welding.
The defective section is removed by grinding, machining, or carbon arc gouging. Gouging must be followed by grinding to remove surface irregularities and any carbon contamination left behind. The excavated groove needs a minimum root radius of 6.3 millimeters and sidewall angles of at least 10 degrees on each side to allow proper access for re-welding. Before any new weld metal goes in, the excavation is inspected with magnetic particle testing to confirm that all defective material has been removed.
Re-welding uses stringer bead technique, with each pass carefully placed to avoid stacking start and stop points in the same location. The repair weld is built up with at least 3 millimeters of overfill, then ground flush to match the original plate surface. Final inspection with radiographic testing is performed no sooner than 48 hours after the repair has cooled to ambient temperature, allowing time for any delayed cracking to develop before the weld is cleared.