Common MIG Welding Defects and How to Fix Them

Every MIG welder, from a first-week apprentice to a shop veteran running six-figure fabrication contracts, eventually stops mid-pass and stares at a bead that just looks wrong. The arc sounded steady, the wire fed smoothly, and the joint still came out porous, undercut, or cracked at the toe. MIG welding defects rarely come from one dramatic mistake. They almost always come from small, compounding issues in parameters, gun technique, shielding gas coverage, or material preparation that stack up until the weld fails to look — or perform — the way it should.

This guide works through the defects that show up most often in gas metal arc welding (GMAW, the formal name for MIG), explains why each one happens, and lays out the specific adjustments that fix it. It also covers the parts of the process that most troubleshooting lists skip: how to tell a cosmetic flaw from a structural one, how wire feed and equipment problems masquerade as technique problems, and when the right answer is to stop, grind it out, and reweld rather than patch over it.

What Counts as a Defect vs. a Cosmetic Imperfection

Not every rough-looking bead is a defect in the technical sense, and conflating the two leads to either wasted rework or, worse, accepted welds that shouldn't be. International and industry standards draw a clear line here. ISO 6520-1 catalogs weld imperfections into six groups — cracks, cavities such as porosity, solid inclusions, lack of fusion or penetration, imperfect shape (undercut, overlap, misalignment), and miscellaneous issues like spatter or arc strikes. ISO 5817 then sets dimensional acceptance limits for those imperfections across three quality levels, B, C, and D, from stringent to moderate.

The practical takeaway: every irregularity is an imperfection, but it only becomes a defect once it exceeds the acceptance limit for the quality level your application requires. A 0.3 mm undercut on a 10 mm structural plate might be perfectly acceptable on a general fabrication job and rejectable on a fatigue-critical one. Cracks and lack of fusion are the exception — codes such as AWS D1.1 and ISO 5817 treat them as unacceptable at any quality level, because both compromise load path integrity in ways that are hard to predict or inspect visually.

Knowing this distinction matters before you reach for a grinder. It tells you which flaws genuinely threaten the joint and which ones are worth improving for the next pass but don't require cutting out today's work.

Why MIG Welding Defects Happen: Four Root-Cause Categories

Nearly every defect covered below traces back to one of four overlapping causes. Keeping these categories in mind turns troubleshooting from guesswork into a process of elimination.

• Parameters — voltage, wire feed speed (which sets amperage), and travel speed out of balance with material thickness and joint type.
• Technique — gun angle, contact-tip-to-work distance (stickout), arc length, and travel consistency.
• Shielding gas — wrong gas type for the base metal, incorrect flow rate, or gas coverage disrupted by wind, drafts, or a dirty nozzle.
• Preparation and equipment — surface contamination (rust, mill scale, oil, paint), moisture-contaminated wire, worn liners or contact tips, and poor ground/work clamp connections.

Most real-world defects involve more than one category at once. A bead with both porosity and spatter, for instance, often points to gas coverage problems compounding with parameters that are already too hot. Working through causes in this order — gas and cleanliness first, then parameters, then technique — resolves most issues faster than chasing symptoms individually.

Porosity: Trapped Gas in the Weld Metal

Porosity is small cavities or pinholes in the weld bead, caused by gas becoming trapped in the molten pool before it solidifies. It's arguably the most common MIG defect because so many different upstream problems can produce it.

Common causes:

• Base metal contamination — oil, rust, mill scale, paint, or galvanized coatings introducing gas-forming compounds into the pool
• Inadequate shielding gas coverage from low flow, a leaking hose, a clogged or spatter-caked nozzle, or drafts and wind blowing the gas envelope away
• Counterintuitively, too much gas flow, which creates turbulence that pulls atmospheric air back into the shielding zone
• Wire that has absorbed moisture from humid storage, introducing hydrogen into the pool
• Excessive stickout, which extends the unshielded travel distance the wire has to cover before entering the arc

How to fix it:

Start with cleanliness — grind or solvent-clean the joint back to bare metal before welding. Check gas flow at the regulator, keeping it in the commonly recommended range of roughly 20–25 CFH (9–12 L/min) for typical shop conditions, and increase it modestly if welding near open doors or fans rather than assuming more is always safer. Inspect the gas line, MIG gun, and nozzle for leaks, cracks, or spatter buildup, and keep the contact-tip-to-work distance around 3/8 to 1/2 inch (10–12 mm). Store wire spools in a dry, temperature-stable area, and discard wire that shows surface rust or has sat opened in humid conditions for long periods.

Excessive Spatter

Spatter is the fine scatter of molten metal droplets thrown from the arc that stick to the surrounding base metal. A small amount is normal in GMAW; heavy spatter signals an unstable arc and usually means wasted filler metal, extra cleanup time, and sometimes a sign of deeper parameter problems.

Common causes:

• Voltage or wire feed speed set too high for the material and wire diameter
• Excessive stickout, which lengthens the arc and destabilizes droplet transfer
• Contaminated base metal or the wrong shielding gas for the transfer mode — for example, high-CO₂ mixes tend to spatter more than argon-rich blends in spray transfer
• Worn contact tips that disrupt current flow and wire alignment

How to fix it:

Dial voltage and wire feed speed back toward the middle of the manufacturer's recommended chart for the wire diameter and material thickness, then fine-tune from there rather than guessing at extremes. Shorten stickout to roughly 3/8 inch, clean the base metal thoroughly, and replace contact tips at the first sign of wear or an oversized bore. Anti-spatter spray on the nozzle and surrounding metal doesn't prevent spatter but makes cleanup much faster.

Undercut

Undercut is a groove melted into the base metal at the toe of the weld that the filler metal fails to fill back in. It thins the base material right where stress tends to concentrate, which is why codes are strict about it on structural and fatigue-loaded joints.

Common causes:

• Excessive voltage or amperage relative to travel speed, melting more base metal than the puddle can fill
• Travel speed too fast, so the arc moves on before the pool levels out
• Poor gun angle, favoring one side of the joint and washing out the far toe
• Arc length too long, spreading heat over a wider area than the puddle can absorb

How to fix it:

Reduce voltage or wire feed speed slightly, and slow travel speed enough to let the puddle fill the melted edge before the arc moves past it. Keep gun angle close to the joint centerline — a 0 to 15 degree push or drag angle is the general starting point most trainers teach — and keep the arc length as short as consistently possible. On fillet welds, a slight pause or weave at each toe helps the pool wet out fully before continuing.

Burn-Through

Burn-through is complete penetration through the base metal, leaving a hole rather than a controlled weld pool. It shows up most often on thin sheet, typically under about 1/8 inch (3 mm, roughly 12 gauge), where the margin between "just enough" and "too much" heat is narrow.

Common causes:

• Heat input too high for the material thickness, from excessive voltage, wire feed speed, or both
• Travel speed too slow, letting heat build up in one spot
• Poor joint fit-up, with gaps that concentrate heat and offer nothing for the puddle to bridge

How to fix it:

Drop voltage and wire feed speed to settings appropriate for thin-gauge material, and increase travel speed so the arc doesn't dwell in one place. Consider a smaller-diameter wire, which allows lower amperage while maintaining stable arc characteristics. On thin sheet, stitch or skip welding — short weld segments with cooling gaps between them — controls heat buildup far better than one continuous pass, and it's a standard technique for reducing warping as well as burn-through.

Lack of Fusion and Cold Lap

Lack of fusion happens when the weld metal fails to bond completely with the base metal or with a previous pass. Cold lap (also called overlap) is a closely related surface-level version of the same problem, where filler metal rolls over the base metal at the toe without actually fusing to it. Both can look acceptable from the surface while leaving a real gap underneath — which is exactly why codes treat lack of fusion as unacceptable at any quality level rather than something to size-limit like porosity.

Common causes:

• Insufficient heat input, often from voltage or wire feed speed set too low
• Wire feed speed too high relative to voltage, which crowds filler metal into the puddle faster than it can properly wet the base metal
• Incorrect gun angle or travel speed that keeps the arc from reaching the joint's sidewalls
• Contaminated or oxidized base metal preventing proper wetting
• Riding on top of the puddle rather than keeping the arc on its leading edge

How to fix it:

Increase voltage and wire feed speed enough to fully melt both the base metal and the previous bead, and keep the arc on the leading edge of the molten pool rather than letting the puddle get ahead of the arc. Slow travel speed slightly to give the joint sidewalls time to fuse, and confirm the gun angle directs heat into the joint rather than along the surface. This is one of the few defects where cleaning alone won't fix a parameter problem — contamination and cold settings often occur together, so address both.

Incomplete Penetration

Incomplete, or lack of, penetration means the weld doesn't extend fully through the joint thickness or into the root, leaving an unfused section beneath the visible bead. It's closely related to lack of fusion but specifically concerns depth into the joint rather than bonding at the surface.

Common causes:

• Insufficient amperage or travel speed too fast to allow adequate heat into the root
• Excessive stickout, reducing the effective heat delivered to the joint
• Joint design with too narrow a root opening or too large a root face for the process

How to fix it:

Increase wire feed speed (and therefore amperage) to match joint thickness, and shorten stickout to improve heat transfer. Slow travel speed enough for the arc to reach the joint root, and where joint design allows, open the root gap slightly or reduce the root face to give the arc a clearer path to full penetration. On thicker material, multiple passes with proper interpass cleaning outperform trying to force full penetration in a single pass.

Cracking: Hot Cracks and Cold Cracks

Cracking is the defect category with the least tolerance for error, because cracks propagate under load and are never acceptable under any weld quality level. MIG welding produces two distinct crack types with different causes and different fixes.

Hot cracks form during or immediately after welding, while the metal is still above roughly 1000°F, and they primarily affect the weld metal itself. They're driven by solidification stresses combined with low-melting-point elements such as sulfur segregating at grain boundaries as the weld cools. Fast travel speed, narrow deep beads, and highly restrained joints all increase the risk.

Cold cracks, sometimes called hydrogen-assisted or delayed cracks, form below about 200°C (392°F) and can appear hours, days, or even weeks after welding — which is what makes them dangerous in inspection terms. They require three conditions at once: dissolved hydrogen in the weld, a hardened or sensitive microstructure, and residual stress. Hydrogen can enter the weld from rust, oil, or moisture on the base metal, from damp wire, from the shielding gas, or simply from atmospheric humidity.

How to fix and prevent cracking:

For hot cracks, slow travel speed to produce a wider, shallower bead profile rather than a narrow, deep one, reduce joint restraint where the design allows, and use filler metal matched to the base metal's chemistry rather than a mismatched substitute. For cold cracks, the priority is controlling hydrogen and cooling rate: clean the joint thoroughly, keep wire and flux dry, and preheat thicker or higher-carbon steel sections so the joint cools more slowly, giving trapped hydrogen time to diffuse out before the microstructure hardens. On critical structural work, post-weld heat treatment or a documented delay before inspection is standard practice specifically because of this delayed-cracking behavior.

Distortion and Warping

Distortion isn't a localized flaw in the bead itself — it's a dimensional problem in the finished part, caused by uneven heating and cooling pulling the material out of shape. It's especially common on thin sheet metal, which heats quickly and has little surrounding mass to absorb and dissipate that heat evenly.

Common causes:

• Excessive heat input concentrated in one area or one continuous long pass
• Welding only one side of a joint, so contraction pulls unevenly
• Poor fixturing, allowing the part to move as it heats and cools
• Oversized welds carrying far more filler metal than the joint actually requires

How to fix and prevent it:

Use stitch or skip welding on long joints — short weld segments with cooling gaps between them — which can cut total heat input dramatically compared with one continuous pass. Weld on alternating sides of a joint (backstepping) to balance contraction forces, and clamp or fixture the workpiece to resist movement while it heats and cools. Size the weld to the joint's actual structural requirement rather than overwelding, since larger welds mean more shrinkage force. On assemblies that are already warping, preheating the surrounding area can reduce the temperature differential driving the distortion.

Wire Feed and Arc Instability Issues

A category of "defects" that trips up beginners and experienced welders alike doesn't originate in the weld pool at all — it starts at the wire feeder and gun. These issues often produce symptoms that look like parameter problems but actually need a mechanical fix.

• Birdnesting — wire tangling at the drive rolls, usually from excessive drive roll tension, a clogged liner, or resistance further down the gun caused by a worn or oversized contact tip.
• Burnback — the wire fuses to the contact tip instead of arcing cleanly, often from wire feed speed set too low relative to voltage, or from a worn tip with too much clearance around the wire.
• Stubbing — the wire jabs into the base metal without striking a stable arc, typically from wire feed speed too high relative to voltage.
• Erratic arc / hunting — the arc wanders or surges, frequently traced to a poor ground clamp connection, inconsistent stickout, or a liner packed with debris.

How to fix it:

Work through the feed path methodically: check drive roll tension and groove size against the wire diameter, inspect the liner for kinks or debris, and replace contact tips showing an enlarged or oval bore. Confirm the ground clamp has clean, solid contact with bare metal — a corroded or loosely clamped ground causes more mystery arc problems than most welders expect. If instability appears only at certain settings, cross-check those settings against the wire manufacturer's chart rather than assuming the machine itself is at fault.

Quick Diagnostic Table

Symptom	Most likely cause	First adjustment to try
Small holes or pinholes in bead	Lost gas coverage or contamination	Check gas flow, hoses, nozzle; clean base metal
Heavy spatter around the bead	Voltage/WFS too high, long stickout	Reduce settings; shorten stickout to ~3/8 in
Groove at the weld toe	Too much heat, travel too fast	Slow travel; reduce voltage; adjust gun angle
Hole burned through material	Heat input too high for thickness	Lower settings; increase travel speed; stitch weld
Bead sits on top without bonding	Heat too low, dirty surface, wrong WFS/voltage balance	Increase heat input; clean joint; keep arc on leading edge
Weld doesn't reach joint root	Amperage too low, travel too fast	Increase WFS/amperage; shorten stickout
Crack appearing during cooling	Fast travel, high restraint, sulfur-rich base metal	Widen bead profile; reduce restraint
Crack appearing hours or days later	Hydrogen contamination, hard microstructure	Preheat; dry wire; clean joint thoroughly
Part bows or twists after welding	Uneven, concentrated heat input	Stitch weld; backstep; fixture the part
Wire tangling at the feeder	Feed path resistance or tension	Check drive rolls, liner, contact tip

Building a Defect-Prevention Routine

Most defects above share upstream causes that a short pre-weld routine catches before they become rework:

• Clean the joint to bare metal — grinding, wire brushing, or solvent degreasing as appropriate for the contamination present
• Confirm gas type and flow rate match the process and material, and inspect hoses, fittings, and the nozzle for leaks or spatter buildup
• Set voltage and wire feed speed from the wire manufacturer's chart for the material thickness and wire diameter, then fine-tune on scrap
• Check contact tip condition and stickout before starting a production run, not after spatter or porosity shows up
• Verify the ground clamp has clean, solid contact directly on or near the workpiece
• Store wire in a dry, covered space and avoid leaving spools open in humid shop conditions

None of these steps takes more than a few minutes, and together they prevent the majority of defects covered in this guide before the first arc is struck.

When a Defect Is Serious Enough to Reweld

Cosmetic irregularities — light spatter, minor surface roughness, a slightly uneven bead profile — are worth improving for technique's sake but don't automatically require rework. Cracks and lack of fusion are the opposite case: because both are excluded at every ISO 5817 quality level and every major structural code, they should be ground out and rewelded regardless of how the application seems to tolerate them, since their real risk shows up under cyclic loading or over time rather than at inspection.

Porosity, undercut, and incomplete penetration fall in between. Whether they need correction depends on their size relative to material thickness and the quality level the application actually requires — a decorative bracket and a pressure vessel nozzle don't carry the same tolerance for a 1 mm surface pore. When in doubt on a structural, pressure-bearing, or safety-related joint, treat the applicable code's acceptance table as the answer rather than a visual judgment call.

Getting Consistent Results

Most MIG welding defects trace back to a handful of variables: heat input relative to material thickness, shielding gas integrity, joint cleanliness, and the mechanical condition of the gun and feeder. Learning to read a bad bead — is it too hot, too cold, contaminated, or unshielded — turns troubleshooting from trial and error into a short, repeatable checklist. Combine that diagnostic habit with a few minutes of pre-weld preparation, and the majority of the defects covered here stop showing up in the first place. When they do appear, treat the fix as data: a groove at the toe is the metal telling you the heat-to-travel-speed balance is off, not a reason to fight the machine harder.