MIG Welding Faults: The Complete Guide to Diagnosing and Fixing GMAW Defects

Why MIG Welding Produces So Many Faults (And What You’re Really Fighting)

The bead looked textbook under the hood. Steady arc, smooth ripples, no drama. Then the inspector flips up the lens and points to a trail of pinpoint holes that never showed up through the dark glass. The weld passed the eye test and failed the bend test. That gap between how a MIG weld looks and what it really is explains why GMAW is so easy to run badly.

Stick welding gives you constant feedback. If your arc length wanders or the slag starts acting up, you feel it immediately. MIG is quieter, faster, and far less forgiving of small mistakes you can’t see. It hides its errors until the metal cools, because there’s no slag to crack off and expose what went wrong underneath. That delay is the whole problem.

A Process With Too Many Hands on the Wheel

Gas metal arc welding uses a continuously fed consumable electrode with an externally supplied shielding gas, rather than a flux-coated stick electrode (or the self-shielded flux-cored wire used in gasless MIG welding). The molten pool is protected from oxidation and nitrogen contamination by gas flowing over it — usually argon, helium, or CO2 mixtures, depending on the material. Simple in theory, punishing in practice. You’re juggling six or seven variables at once:

• Voltage (arc length)
• Wire feed speed (current/amperage)
• Travel speed
• Contact-tip-to-work distance, or stickout
• Gas flow rate and coverage
• Gun angle and direction
• Base metal condition — cleanliness, thickness, fit-up

None of these lives in isolation. Nudge your stickout and you change the amperage, which changes the penetration, which changes the bead profile. That’s why a MIG fault is almost never bad luck. It’s a symptom of one variable fighting another.

The Four Families of MIG Problems

Lincoln Electric sorts the mess into four broad brackets: weld porosity, improper bead profile, lack of fusion, and faulty wire delivery tied to equipment setup and maintenance. Almost every fault you chase lands in one of those boxes. Porosity is trapped gas. Improper bead profile covers convex, concave, and overlapping beads. Lack of fusion hides at the sidewalls or root where the arc never tied in. The last group points to the machine — the liner, drive rolls, and contact tip.

Discontinuity vs. Defect: Know the Difference

A discontinuity is any interruption in the weld: a pore, a void, an undercut, a crack, a slag inclusion. It doesn’t automatically mean failure. A defect is a discontinuity severe enough to fail the applicable acceptance criteria — a structural code, a customer spec, or an engineering drawing. A small pore might be acceptable on a garden gate and rejectable on a pressure vessel. The line moves with the standard you’re welding to.

Why Your Hands Matter as Much as Your Settings

MIG is a semi-automatic process. The machine feeds wire and gas, but you control travel speed, gun angle, and stickout. Technique carries as much weight as any dial on the front panel. A welder who chases faults by twisting knobs alone only solves half the problem.

This guide walks the whole field. We’ll start with the controls that create or cure every fault, then work through porosity, fusion defects, heat-input failures like burn-through and undercut, bead-profile problems, equipment faults, distortion and cracking, material-specific issues, and finally a practical diagnosis workflow you can run at the bench — much like the troubleshooting approach Miller recommends.

The sooner you stop treating MIG faults as random, the sooner your weld quality stops depending on luck.

MIG Welding Variables Explained: The Controls That Create or Cure Every Fault

Every MIG fault traces back to a handful of variables. Before you start chasing pinholes or a wandering toe line, you need to understand what each control actually does. MIG welding is a balancing act between heat, metal deposition, and gas protection. Change one dial and you shift the other two.

Voltage: Setting the Arc Length

Voltage controls arc length, not current. Turn it up and the arc stretches; turn it down and the arc tightens against the puddle. Too little voltage produces poor arc starts, weak penetration, excessive spatter, and a tall convex bead. Too much gives you a wide, flat, unstable arc that can undercut the edges. Think of voltage as the shape control for your bead profile.

Wire Feed Speed: Your Amperage Dial

On a constant-voltage MIG machine, wire feed speed (WFS) sets amperage. Feed wire faster and you burn more current, depositing more metal. Feed it slower and the arc runs colder. This is why WFS and voltage must be tuned together: voltage for arc length, WFS for heat.

Travel Speed and Contact-Tip-to-Work Distance

Travel speed controls heat input per inch. Move too fast and you get a narrow, cold, ropey bead with poor fusion. Move too slow and you pile up metal, overheat the part, and risk burn-through. Contact-tip-to-work distance (CTWD) matters just as much. Push the gun too far from the work and resistance rises, the arc wanders, and gas coverage collapses. Miller recommends extending the wire no more than 1/2 inch past the nozzle for good gas coverage. A practical standoff of roughly 1/4 to 1/2 inch keeps your arc stable and your gas blanket intact. Worn contact tips throw this off too, so check out Contact Tip Welding when your settings suddenly stop behaving.

Gun Angle and Travel Direction

Drag (pulling) gives you more penetration and a flatter bead. Push (pushing) gives less penetration but better visibility and a cleaner look. A 10- to 15-degree work angle off vertical is the usual starting point. Consistency here is half the battle; sloppy angle changes create spatter and uneven fusion.

Shielding Gas: Selection and Flow Rate

Gas shields the molten puddle from the atmosphere. Common GMAW shielding gas flow rates run roughly 20-50 CFH. Too little and you get porosity; too much and you create turbulence that pulls air into the arc. On steel, mix choice changes everything. A 75% argon/25% CO2 mixture typically produces less consistent transfer and more spatter than a 90% argon/10% CO2 mixture. Less CO2 means a smoother, calmer arc.

Metal Transfer Modes

The way metal crosses the arc is the last piece of the puzzle. Transfer mode decides whether you’re welding in a stable spray or a spattery short circuit, and it depends on voltage, gas, and wire size together.

• Short-circuit — low heat, thin metal. The wire touches the puddle and shorts out. Short-circuit transfer always creates higher spatter levels than spray transfer.
• Globular — large, irregular drops. Rough arc, more spatter, mostly a sign your settings are off.
• Spray — high heat and high voltage, with a fine, axial spray. Smooth, deep-penetrating, best on thick steel.
• Pulsed spray — spray transfer at lower average heat. Great for thinner material while keeping that smooth arc.

A good starting point for 1/8-inch mild steel is around 18-19 volts and 200-220 inches per minute of wire feed, but treat that as a rough launch pad rather than gospel. Machine brands, wire diameter, and joint type all shift the numbers. When a weld misbehaves, adjust one variable at a time and watch what changes. Manufacturers publish detailed fault charts you can cross-check; Miller’s troubleshooting guide and Lincoln’s problems-and-remedies page are both worth bookmarking.

Porosity: The Most Common MIG Welding Fault and How to Eliminate It

Porosity is gas trapped in solidifying weld metal. On the surface it shows up as pinholes scattered in the bead; underneath it hides as internal cavities you only find with X-ray, ultrasonic testing, or a bend test. Either way, the metal you deposited isn’t solid steel — it’s a sponge. Every pore removes load-bearing area, concentrates stress, and gives a crack a place to start. That’s why porosity is the single most common MIG fault and the one that fails inspections most often. If your beads look suspect, compare them against Good and Bad MIG Welds: How to Tell the Difference before you start tearing the machine apart.

Surface pinholes are the ones you can see. Subsurface porosity is the dangerous kind, because a weld can pass a visual check and still fail a bend or X-ray test. Left unchecked, it can fail an entire production run before anyone catches it. Porosity almost never comes from one mystery cause — it comes from three buckets: lost shielding gas coverage, contaminated metal, and bad gas. Work through them in order.

Bucket 1: Shielding Gas Coverage Failures

MIG relies on a blanket of shielding gas to keep the atmosphere away from the molten pool. Nitrogen and oxygen from the air dissolve into the weld and come back out as bubbles as it freezes. Anything that thins or breaks that blanket causes porosity.

• Flow too low or too high. Too little gas leaves the arc exposed. Too much creates turbulence that sucks air into the stream. Most solid-wire GMAW runs comfortably around 20 to 30 CFH, tuned to conditions.
• Leaks. A cracked gas hose, worn O-rings, a loose fitting, or a broken gas solenoid pulls atmosphere straight into the shielding stream. This is the classic cause of “porosity all shift, then it clears” — a leak you can’t see but the arc can.
• Spatter-clogged nozzle or plugged diffuser. Once the nozzle or the diffuser’s gas ports fill with spatter, the gas comes out crooked. Keep both clean.
• Drafts. Open shop doors, fans, and even a strong HVAC vent will blow the gas away.
• Gun angle and stickout. Too much stickout or a drag angle pushes the gas blanket behind the arc instead of over it.

Bucket 2: Base Metal and Wire Contamination

Oil, grease, rust, mill scale, paint, and moisture on the plate all release gas as the pool solidifies. That gas has nowhere to go but up through the bead. Clean the joint back to bare, dry metal before you strike an arc, and wipe the wire and plate down. For a deeper look at getting your prep and technique right, see our guide on Welding Practice.

Bucket 3: Gas Quality

A wet or contaminated cylinder — or one that sat with the valve open — puts moisture into the line and into the weld. Swap the cylinder and purge the hose if you suspect it. Also confirm you’re running the gas mix your procedure calls for; a wrong or stale blend behaves like contamination.

Fixes That Actually Work

1. Set flow correctly and verify it at the nozzle with a flowmeter, not the regulator gauge alone.
2. Leak-test the whole path: hose, gun, O-rings, fittings, solenoid.
3. Clean the base metal and the wire. Remove rust, scale, oil, and moisture.
4. Use a large enough nozzle and a threaded-in gas diffuser so flow stays laminar. Replace worn contact tips — see Contact Tip Welding for wear signs.
5. Clean the nozzle and diffuser — spatter kills coverage.
6. Weld with a push (forehand) angle so the gas lays ahead of the arc.
7. Screen the area so wind stays at 5 mph or less.
8. Match filler to the base metal. Deoxidizers like silicon and manganese in the wire combine with oxygen instead of letting it form pores.

That’s the full chain: gas coverage, clean metal, clean gas. Get all three right and porosity mostly disappears.

Lack of Fusion, Incomplete Fusion and Lack of Penetration: The Hidden Strength Killers

A weld can look clean, sound, and perfectly dressed while its fusion boundaries are nothing more than two pieces of metal sitting next to each other. That’s the danger of lack of fusion and lack of penetration. The bead cools, the toes look tight, the crown is nicely rippled, and the joint is quietly weak. Send it to ultrasonic testing or cut a coupon through the middle, and the truth falls out.

The two faults are related but not the same. Lack of fusion means the deposited weld metal fails to bond to the sidewall or to a previous pass. The filler sits in the groove, but the grains never knit together, leaving a mechanical seam instead of a metallurgical bond. Lack of penetration is a root problem: the arc never melted deep enough into the base metal at the joint root, so the plates are joined only near the top. One is a bond failure along the vertical faces; the other is a depth failure at the bottom. Both cut the effective throat, and both count as serious discontinuities in most codes.

Why They Hide So Well

These defects live beneath the surface, which is exactly why they slip past visual inspection. A surface-breaking overlap welding defect at least gives you a toe line to see. Lack of fusion gives you almost nothing. The reliable ways to find it are destructive testing, ultrasonic inspection, or radiographic methods that see inside the joint. That’s why fabricators cutting coupons for procedure qualification often uncover fusion problems their welders never suspected.

What Causes Cold Lap and Incomplete Fusion

Welding current that is too low is the most common cause of lack of fusion in MIG welding. Running cold means the arc doesn’t carry enough energy to melt both the filler and the sidewall at once, so the puddle rides on top of cold base metal instead of digging in. The usual lineup:

• Amperage too low, often from a conservative voltage/wire speed setting.
• Travel speed too fast, which produces a narrow, convex bead with poor tie-in at the toes and insufficient penetration.
• Wrong gun angle, especially a steep drag angle that pushes metal ahead of the arc.
• Wire diameter mismatched to the joint, or excessive stickout that burns off wire speed before the arc reaches the metal.
• Poor joint fit-up with a gap or misalignment the puddle can’t bridge cleanly.
• Mill scale, oxide, and contamination on the fusion faces, which can prevent bonding even when heat input is adequate.

That last one catches people out. You can be running hot enough to fuse, but a heavy layer of mill scale or aluminum oxide sits between the puddle and the sidewall like a barrier, and the metal never truly wets in. Cleanliness on the fusion faces matters as much as the settings.

How to Fix It at the Machine

Start by raising amperage until the puddle wets out and ties into both toes. Spray transfer usually helps prevent lack of fusion because the arc delivers a steady, high-energy stream that melts the sidewalls, so consider switching out of short-circuit where thickness allows. Place the arc at the leading edge of the puddle rather than on top of deposited metal, and watch the toes to confirm they’re washing in rather than sitting proud.

Then slow down. Reduced travel speed lets heat soak into the joint and improves tie-in. Correct the joint preparation so you have clean, properly sized faces with the right gap. On stainless, tungsten and silicon can promote cold lap, a shallow defect where the puddle rides over the surface without bonding, so filler selection and travel technique both matter. Aluminum adds its own twist: oxide removal and polarity change how the arc cleans and fuses, which is why DC welding aluminium behaves so differently from AC. Get the heat, angle, travel speed, and cleanliness right, and fusion stops being a hidden gamble.

Burn-Through, Undercut and Crater Faults: Heat Input Failures

These three defects share one root cause: too much heat entered the joint, or the operator mishandled the moment the arc stopped. On thin material they appear almost instantly, which is why anyone welding sheet metal fights them constantly.

Burn-Through: When the Arc Wins

Burn-through happens when weld metal penetrates completely through the base material and leaves a hole. It is especially common on stock below 1/8 inch (roughly 12 gauge), and on aluminum, which conducts heat about six times faster than steel while melting at a much lower temperature. That combination makes aluminum highly susceptible to warping and blow-through.

Excessive heat input usually traces back to one of these settings or habits:

• Voltage too high — a long, hot arc melts more base metal than the joint needs.
• Wire feed speed too high — current climbs and the puddle grows faster than you can travel.
• Travel speed too slow — dwelling pours heat into one spot until it collapses.
• Root gap too wide — nothing behind the puddle supports the molten metal.
• Pausing or hesitating — a stationary torch burns straight through.

Fixes mirror the causes: lower voltage and wire speed, move faster with a steady rhythm, tighten fit-up, and if you must pause, shift the arc onto thicker material. On very thin sheet, drop to a smaller wire diameter and try stitch or backstep techniques. If a hole opens anyway, stop, let it cool, and repair from both sides instead of chasing the puddle.

Undercut: The Groove Along the Toe

Undercut melts a groove into the base metal along the weld toe. The joint keeps its width but loses cross-section, reducing the effective throat thickness and creating a stress riser that can start a crack under load. It is a close cousin of overlap, except the metal is gouged away rather than piled up.

The usual culprits are excessive voltage or current, an arc that is too long, a torch angle leaning too far from vertical, and travel speed pushed too fast for the puddle to fill in behind. The remedy: shorten the arc, trim voltage slightly, slow down just enough for filler to flood the toe, and correct your work and travel angles so the arc points into the joint rather than at the plate edge. Where undercut already exists, grind it out and run a light pass to restore the profile. Running a second overlapping pass is a common repair, though it blurs the line between technique and defect — our guide to overlapping welding untangles that.

Craters and Crater Cracks at Termination

A crater forms at the end of a weld where the molten pool shrinks and solidifies last. If you release the trigger and walk away, that shrinking pool can pull itself apart, leaving a crater crack. The fix is to fill the crater before breaking the arc: pause, add a beat of wire, then reverse travel briefly to backfill, or run the weld onto a run-off tab. Many machines include a crater-fill or burn-back setting that tapers current and wire feed automatically. This is the same controlled-termination discipline covered in our welding practice guide.

For a fast field diagnosis, picture a simple three-column remedy table: symptom in the first column, likely heat-input cause in the second, and the single adjustment to make in the third. Burn-through maps to voltage or wire speed too high and travel too slow, undercut to an arc too long or travel too fast, and crater cracks to an arc terminated before the pool filled. Reading down that table turns guesswork into a sequence of one-change-at-a-time tests.

Heat is the shared thread. The weld tells you when it has had enough — a widening puddle, an edge starting to sag, a toe line cutting into the plate. Trust those signals, ease off, and these faults largely disappear.

Spatter, Convex and Concave Beads: Diagnosing Bead Profile Faults

Spatter and odd bead shapes are often waved off as cosmetic. They are not. Spatter is molten metal flung onto the workpiece and bench around it, and bead profile is the clearest picture you get of how balanced your arc really is. Read both properly and you can correct the weld before a cosmetic issue becomes a strength problem.

Spatter: More Than a Cleanup Problem

Spatter is the peppering of tiny metal balls welded to the surface beside your bead. It costs you grinding and wire-brush time on every part, and worse, it can hide porosity, undercut, or contamination sitting underneath. Common causes include:

• Voltage too low relative to wire feed speed — by far the most common contributor, regardless of transfer mode
• Contaminated base metal: mill scale, rust, oil, paint, or moisture
• Wrong torch angle, or pushing when the wire calls for dragging
• Excessive stickout that lets the arc wander
• Insufficient shielding gas flow or a blocked nozzle
• A wrong-size, worn, or incorrectly recessed contact tip
• Pure CO2 shielding gas, which runs harsher than argon mixes

Poor arc starts and a ropey, cold bead usually share that same low-voltage root cause. If the toes of the weld will not wet in, you are running too cold. For a deeper look at how the arc behaves on a properly dialled machine, see our guide to good and bad MIG welds.

Cleanup is where spatter really stings. Depending on how the weld is finished, every spatter ball has to be chipped, ground, or wire-brushed away before paint or powder coat. Worse, spatter sitting over a weld can trap the very porosity and contamination you are trying to find, so you clean a bead that was never sound to begin with.

Convex vs. Concave: What Bead Shape Tells You

Convex beads sit high and narrow with steep sides that fail to tie into the parent metal. That is a cold weld: too little voltage, travel too fast, or both. The toes never fuse, so the joint looks finished but lacks penetration.

Concave beads are the opposite — wide, shallow, and dished. Voltage is too high, which flattens the bead, thins the throat, and invites centerline cracking. A slightly convex-to-flat profile with smooth, blended toes is what you are aiming for. Excessive reinforcement is just a convex bead taken too far; underfill is a concave one that has been over-widened.

A wandering toe line, a bead that sits off-centre, or reinforcement that varies along the length all point to inconsistent travel speed, stickout, or torch angle rather than a machine fault.

How to Cut Spatter at the Machine

• Switch to spray transfer on thicker steel where the machine allows it — the arc is far smoother than short-circuit.
• Try metal-cored wire. All other things being equal, it produces a smoother arc and less spatter than solid wire.
• Match polarity to the wire. With self-shielded flux-cored, welding straight polarity (electrode negative) and dragging minimises spatter — see polarity for flux core welding for the full breakdown.
• Fit the correct contact tip size and replace worn tips regularly. A worn or wrong contact tip alone will throw spatter.
• Clean the base metal and set the right gas mix and flow.

Not sure whether that wire is even a MIG wire? Read Is flux core the same as MIG? to avoid chasing a fault that is really a process mismatch. Once your profile and spatter are clean, you have eliminated a whole branch of wire welder problems. It is the fastest, cheapest fault-finding tool you own.

Wire Feeding and Equipment Faults: When the Machine Is the Problem

Not every bad bead is your fault. MIG is a machine-driven process, and when the wire, tip, gas, or ground path misbehaves, the arc does too. If your technique is sound on the same settings that worked yesterday, suspect the equipment. These are the faults that live inside the gun, the feeder, and the cables.

Wire Feed Problems

Faulty wire delivery is the fourth major category of MIG trouble, and it usually announces itself as a stuttering arc and an uneven bead.

• Drive roll tension. Too loose, and the roll slips, so wire speed drops and the arc sputters. Too tight, and the wire deforms, kinks, and bird-nests at the feeder, a classic problem manufacturers list in their MIG troubleshooting guides. Set tension just until the wire stops slipping when you pinch it.
• Spool drag and brake tension. If the spool overruns, wire unspools and tangles; too much brake drag, and the feeder fights the spool, causing slip and surge.
• Wrong drive roll groove size. Use the groove matched to your wire diameter and type — V-groove for solid, U or knurled for flux-cored.
• Worn or kinked liner. A dirty or kinked liner creates drag, changing wire feed speed and destabilizing the arc. Blow it out or replace it.
• Excessive gun cable length or bend. Long, tightly coiled cables add friction. Keep the lead as straight as possible.
• Wire diameter and stickout. Excessive stickout adds resistance, and the wrong wire size for a liner or tip will drag or wobble.

Contact Tip Faults

The contact tip is where current transfers to the wire, so its condition decides arc stability. A wrong-size, worn, or badly recessed tip can restrict or loosen the wire path, causing spatter and porosity. An oversized bore lets the wire wander, so the arc scatters. A bore packed with spatter pinches the wire and causes feed hesitation. And a tip recessed too far inside the nozzle starves shielding coverage. Check our contact tip welding guide for sizing, and replace tips as consumables, not when they fail outright.

Arc Blow

An uncontrollable, erratic arc that repeatedly throws high spatter in one area of the part is the typical symptom of arc blow, a fault Miller’s defect guide flags as magnetic in origin. This is a magnetic problem, not a settings mistake. On DC current, magnetic forces build up, and it gets worse with long arcs, at the ends of joints, and on heavy or irregularly shaped parts. Shorten the arc, reduce current, move the work lead connection point, or switch to AC where the process allows. Repositioning the ground clamp often helps, which is why good quick connect welding leads that let you move fast pay off.

Gas Delivery Equipment Faults

Shielding gas faults look exactly like porosity. Regulator creep pushes flow too high and creates turbulence, or drops it below the rate you need. Leaks at the hose, solenoid, or fittings draw in air. A contaminated cylinder introduces moisture or the wrong gas entirely. The diffuser’s ports can also clog with spatter, creating turbulent or restricted shielding gas flow. Clean the diffuser, check for leaks with soapy water, and verify flow at the nozzle, not just at the regulator.

Shift-Start Maintenance Checklist

Five minutes before you strike an arc prevents most of these faults:

• Inspect the liner and blow it out or replace it if kinked or dirty.
• Replace the contact tip when the bore wears or spatter builds.
• Clean the gas diffuser ports.
• Check drive roll tension and groove size.
• Confirm gas flow and inspect hoses and fittings for leaks.

Learn your machine properly with our beginner’s guide to using a wire welder, and keep a maintenance log so you catch a worn liner or tip before it shows up in a weld.

Distortion, Warping and Cracking: Faults That Grow After the Weld Cools

MIG welding produces clean, fast, forgiving welds, until it doesn’t. Some faults announce themselves the moment you lift the helmet. Others are patient. Distortion twists a finished part minutes after you lay the torch down, and cracks can open hours, days, or even weeks later, sometimes long after the job has shipped and been signed off. These are the failures that slip past a quick visual check and then betray you in service. Because they develop slowly and often in silence, they deserve their own diagnosis, a category apart from the spatter and porosity you can spot the moment the bead cools.

Distortion and Warping: Heat You Can’t Take Back

Every weld shrinks as it cools, and that shrinkage drags the surrounding metal with it. Distortion is not a single mistake but the sum of many small ones: uneven heat input, poor restraint, and a joint that was fighting itself before the arc even struck. Slower travel and higher heat both make it worse, because the surrounding metal absorbs more energy, expands further in the moment, and then contracts harder on the way back down. Thin sheet and aluminium are the worst offenders. They conduct heat away quickly and have little stiffness of their own, so a single slow pass can ripple a panel like a potato chip. Heat is never free; what you put in, you eventually pay back in movement.

Travel speed is the usual villain. A travel speed that is too slow produces a large weld with excessive heat input, resulting in heat distortion and possible burn-through. When the puddle is oversized, most of its energy goes into heating metal that never needed it, and that metal rewards you by moving. Stack poor sequencing, missing clamps, and no backstep technique on top of that, and the part begins to fight itself from both ends. Residual stresses locked in during forming make everything worse, which is why good fit-up is inseparable from metal rolling and welding practice. The joint is not just a gap to fill; it is a set of stresses waiting to be released by your heat.

Distortion rarely announces itself as a crack. It shows up as a bowed flange, a twisted frame, a gap that closed by itself, or a door that no longer latches. Because it happens gradually, operators often blame the material or the part instead of the heat that caused it. The tell is the weld itself: if the bead is wider than it needs to be, the surrounding metal took more heat than the joint required, and the movement you see is simply that excess energy finding its way out.

Hot Cracking and Solidification Cracking

Hot cracking, sometimes called solidification cracking, happens while the weld is still freezing. The metal does not fail because it is cold and brittle; it fails because it is hot and barely holding together. The classic triggers are the wrong filler chemistry, contaminated weld metal, a concave bead with a deep, narrow shape, high-sulfur base metal, and heavy restraint. As the puddle solidifies, sulfur and phosphorus segregate toward the centreline, leaving a thin film of low-melting liquid exactly where the two sides of the weld are trying to pull apart. The last liquid to freeze simply tears, and the fracture often runs straight down the middle of the bead.

High-sulfur base metal is a known cause of weld porosity and can contribute to cracking as well, so free-machining steels deserve special caution. Restraint makes it worse: the stiffer the joint, the more strain the freezing metal must survive. Matching filler chemistry to the base metal and keeping the bead slightly convex both help the weld freeze without tearing. Rounded, generous beads tolerate shrinkage far better than deep, narrow, concave ones, which is why a slightly hotter, wider pass often cracks less than a cold, pinched one. In MIG welding specifically, hot cracking often traces back to a filler that is simply too strong for the base metal, or to a joint that was fitted too tightly. Leave a sensible gap, let the weld shrink as it wants to, and the freezing metal has somewhere to go.

Cold Cracking and Hydrogen Cracking

Cold cracking, also called hydrogen cracking or delayed cracking, is the opposite personality: it waits. It shows up hours after the weld cools, typically below roughly 200°C, and occasionally days later, which is why a job that looked perfect on the bench can fail on the truck. Three things must be present together for it to happen: hydrogen, a hard heat-affected zone, and restraint. Remove any one of the three and the crack never starts. That triangle is the key to prevention, because you can almost always attack at least one corner of it.

When the base metal hardens faster than the filler can keep up, a small heat-affected zone forms with brittle martensite, and trapped hydrogen finishes the job. Hydrogen, often carried in by moisture, oil, rust, or a damp electrode, diffuses into that hardened zone and builds pressure in microscopic voids until the metal splits. That mismatch between filler and base chemistry is why welding hardened steel demands preheat, a matching consumable, and slow cooling. Preheat slows the cooling rate, which gives trapped hydrogen time to escape and keeps the heat-affected zone softer and more forgiving. Crater fill problems and a poor joint configuration also leave cracks where the bead stops, since a starved crater is a stress riser waiting to open.

Moisture is the quiet saboteur here. A damp electrode, a rusty plate, condensation on cold steel, or even humid shop air can deliver enough hydrogen to start the process. Drying consumables and preheating the joint both drive that moisture off before it can do damage.

Fixing Cracks and Catching Them at Inspection

Never weld straight over a crack. This is the single most common field mistake, and it almost always comes back. Cracks cannot simply be welded over; the cracked section, or the entire weld, must normally be removed and redone, because the tip keeps propagating under load. A crack is a stress concentrator: grind or gouge back to sound metal, confirm the tip is gone, and only then rebuild. If you cap a crack without reaching its root, you have hidden the problem, not solved it.

Remember that a cracked weld and a cracked base metal are different animals. Cracking beside the toe of the weld usually points to a heat-affected-zone problem, while cracking down the centreline points to solidification and filler chemistry. Inspection is what catches these faults before they become failures. Visual checks catch surface breaks, dye penetrant reveals fine surface-connected cracks, and ultrasonic testing finds buried ones before they grow. Each method sees something the others miss, so serious work usually combines them rather than trusting any single check. If a crack is found, grind it out completely before any repair pass, and verify the excavation before you start welding again. Catching a crack early is cheap; finding it in service is not.

Prevention: Heat Control Beats Repair

Prevention is mostly a discipline of heat and sequence. None of these steps is glamorous, and none is difficult, but together they stop most distortion and cracking before it ever starts.

• Minimise heat input by raising travel speed and using the lowest amperage that still fuses.
• Weld in a balanced sequence, alternating sides and backstepping long joints.
• Tack, clamp, and brace the part before the first arc.
• Use the filler the base metal actually calls for, not whatever is loaded.
• Control cooling and fill every crater to its full cross-section.

If distortion does appear, resist the urge to press it back cold. A controlled, well-placed heat shrink can pull a part back into line, but brute force usually just moves the stress somewhere else.

Handle these habits early and distortion stays small enough to hammer out, and cracks never get the chance to form. Get them wrong, and you will be cutting out sound-looking welds later. The best time to think about distortion and cracking is before the first arc, never after the part has already moved.

Material-Specific MIG Faults: Aluminum, Stainless, Thin Sheet and Galvanized Steel

The fault framework stays the same, but tricky materials change the numbers. The same settings that lay a clean bead on mild steel can leave you with burn-through, porosity, or a fume cloud on aluminum, stainless, thin sheet, and galvanized steel. Each metal rewrites the rules for heat, gas, and cleanliness, so the diagnosis has to start with the material in front of you.

Aluminum: Fast Heat, Stubborn Oxide

Aluminum conducts heat about six times faster than steel, so heat vanishes into the part almost as fast as you add it. Combine that with a low melting point and you get a metal that warps and burns through with almost no warning. By the time the puddle looks wrong, you are usually already through the workpiece.

The surface makes it worse. Aluminum carries a tenacious oxide layer that melts at a far higher temperature than the metal beneath it, so it must be cleaned mechanically and welded with pure argon shielding. Aluminum wire is also softer than steel wire, with less tensile strength, so it feeds poorly through a standard gun and produces defects without the right equipment. Spool guns, push technique, and higher travel speeds all help manage the heat and keep the wire moving smoothly.

One important safety angle: GMAW of aluminum-magnesium alloys produces relatively high fume rates because magnesium vaporizes easily, and aluminum welding is conducive to ozone production. NIOSH welding guidance treats both as respiratory hazards. If you want to drop the shielding-gas variables, learn how gasless MIG welding works before switching processes, and if a machine lacks AC capability, this breakdown of DC welding aluminium explains the trade-offs.

Stainless: Distortion, Heat Tint and Carbon Pickup

Stainless steel faults cluster around heat. Too much input causes distortion and ugly chromium oxidation, the rainbow heat tint that signals a depleted corrosion-resistant surface. That tint is not just cosmetic; it marks where the chromium has been drawn away, leaving the weld far more likely to rust. Carbon pickup is a related risk when you use the wrong wire or contaminated tooling, since carbon pulled into the weld can seed corrosion later.

Use the correct tri-mix shielding gas for the grade and keep heat low with short, controlled beads. And watch the fume. Stainless welding can generate hexavalent chromium, so OSHA’s welding fact sheet recommends fume capture and proper respiratory protection.

Thin Sheet: The Burn-Through Trap

Thin sheet fails the moment heat input outruns the metal. The classic fixes are smaller wire, lower voltage, and a faster travel speed that moves the arc before it punches through. Fit-up matters too, since gaps invite blow-through. For the fundamentals of holding a stable bead, review this guide to good and bad MIG welds.

Galvanized and Coated Steel: Zinc in the Air

Galvanized steel is coated in zinc, which vaporizes under the arc and creates porosity, spatter, and a dense fume that carries real metal-fume-fever risk. Grind the coating back from the joint, position yourself upwind, and ventilate the space before you strike an arc.

Across all four materials, the pattern holds: control heat input, control the shielding gas, and control what you breathe. Fit proper protective eyewear for welding, then treat cleanliness and ventilation as part of the setup, not an afterthought.

A Practical Fault-Diagnosis Workflow, Prevention and Final Tips

Every fault in this guide leaves a clue. The skill isn’t memorizing fixes — it’s reading symptoms in the right order so you stop chasing ghosts and start fixing the real cause.

1. Look at the Bead First

Before touching a dial, study the finished weld. Is it porous, convex, undercut, off-line? The bead tells you where to start looking. Compare it against reference images of good and bad MIG welds until you can name the defect instantly.

2. Check the Fundamentals in Order

Work through this sequence every time. It moves from the cheapest, most common culprits to the expensive ones.

1. Gas — flow rate, leaks, regulator, correct mix.
2. Polarity — leads in the right sockets for your wire type.
3. Settings — voltage, wire feed speed, and travel relationship.
4. Technique — angle, stickout, travel speed, manipulation.
5. Material — cleanliness, thickness, contamination, fit-up.
6. Equipment — liner, drive rolls, contact tip wear, ground clamp.

Most faults trace back to a break in one of these six links. If the bead looks wrong and the gas is fine, resist the urge to reach straight for the voltage knob.

3. Change One Variable at a Time

This is the single fastest way to isolate a true root cause. Adjust gas, weld, then observe. Move to voltage only after that. Changing three things at once may “fix” the bead while teaching you nothing — and the fault returns next shift.

4. Document the Fix

Keep a log: material, thickness, settings, gas, wire, symptom, and solution. Over months this becomes your personal troubleshooting database, far more useful than any generic chart.

Quick-Reference Symptom Chart

Miller’s MIG troubleshooting guide and Hobart’s wire troubleshooting charts are worth printing for the shop wall, but the table above covers the everyday 80 percent.

Prevention Beats Diagnosis

Most faults are prevented before the arc ever starts:

• Prep and clean. Remove mill scale, oil, paint, and rust; bevel and fit joints properly.
• Store wire dry. Sealed packages, dry cabinet once opened. Moisture means porosity.
• Maintain consumables. Tips, nozzles, liners, and drive rolls wear predictably — replace on schedule, not on failure.
• Follow the WPS. If a procedure is specified, the settings, gas, and filler aren’t suggestions.
• Pre-weld checks. Gas flow, ground, polarity, spool, and drive tension in under a minute.

A clean, well-organized shop makes all of this easier. If you’re setting up or upgrading, see our guide to planning and setting up a safe welding facility.

Safety: Faults Raise the Fume Risk Too

Poor process control isn’t just a quality problem. GMAW generates harmful fumes and gases, and gas-shielded arc welding can produce ozone, nitrogen oxides, and carbon monoxide — worse in confined or poorly ventilated spaces. OSHA’s fact sheet on controlling welding fume emphasizes local exhaust ventilation at the source. Welding fume is classified as a possible human carcinogen by the IARC. Spatter, itself a symptom of bad settings, also burns skin and starts fires. Correct settings cut both risks — and always protect your eyes with proper protective eyewear.

Faults Are Feedback, Not Failure

Every defect is the process telling you exactly which variable is out of balance. Read the bead, work the checklist, change one thing, and write it down. That loop is what separates a welder who fights the same faults for years from one who rarely sees them twice. Skill comes from repetition, but useful repetition comes from deliberate welding practice. MIG has been the backbone of fabrication since its patent-era beginnings for the same reason it’s forgiving to learn and revealing to master: it rewards control.

Frequently Asked Questions

What is the most common MIG welding fault?

Porosity is widely considered the most common MIG welding defect. It is usually caused by inadequate shielding gas coverage or contamination on the base metal, and it shows up as pinholes on the bead or hidden cavities inside the weld.

Why does my MIG weld have tiny holes in it?

Those holes are porosity. The usual culprits are low or turbulent shielding gas flow, a clogged nozzle or diffuser, drafts blowing the gas away, or oil, rust, paint and moisture on the metal releasing gas into the pool. Clean the material and restore proper gas coverage.

What causes excessive spatter in MIG welding?

The most common cause is voltage that is too low relative to wire feed speed, which destabilizes the arc. Dirty base metal, wrong torch angle, too much stickout, a worn contact tip, and pure CO2 shielding gas also increase spatter.

How do I stop burning through thin metal with MIG?

Reduce voltage and wire feed speed, increase travel speed, and switch to a smaller wire diameter. You should also close up the root gap, clamp the parts for heat sinking, and avoid pausing in one spot, since thin sheet and aluminum are the most susceptible.

What is the difference between lack of fusion and lack of penetration?

Lack of fusion is a failure to bond between the weld metal and the sidewall or a previous pass. Lack of penetration is a failure to fuse into the root of the joint. Both are hidden strength-reducing faults, and both are commonly caused by running too cold.

Can I fix a cracked MIG weld by welding over it?

No. Cracks must be removed, not covered. You should grind out the cracked section (or the entire weld) and re-weld it, because welding over a crack usually just hides it and lets it propagate.

Why does my MIG wire keep bird-nesting at the drive rolls?

Bird-nesting usually means drive roll tension is too tight, the liner is kinked or worn, the contact tip is clogged, or the gun cable has too sharp a bend. Check tension first, then inspect the liner and tip.

How much shielding gas flow should I use for MIG welding?

Most GMAW setups run roughly 20 to 50 cubic feet per hour. Too little flow exposes the pool to air and causes porosity, while too much flow creates turbulence that can also pull in atmosphere.

Does MIG welding produce harmful fumes?

Yes. All arc welding produces fumes and gases, and MIG welding of aluminum and stainless can generate ozone, nitrogen oxides, and metal fumes including manganese. Use local exhaust ventilation and appropriate respiratory protection, and never weld in a confined space without controls.

Authoritative Sources & Further Reading

• https://www.millerwelds.com/en-us/resources/knowledge-hub/mig-welding/how-to/troubleshooting-weld-defects
• https://www.millerwelds.com/en-US/resources/knowledge-hub/mig-welding/how-to/the-most-common-mig-weld-defects-on-aluminum-and-steel-and-how-to-avoid-them
• https://www.hobartbrothers.com/resources/technical-articles/mig-troubleshooting-for-metal-cored-and-solid-wire
• https://www.lincolnelectric.com/en-ca/welding-and-cutting-resource-center/process-and-theory/mig-problems-and-remedies
• https://www.osha.gov/sites/default/files/publications/OSHA_FS-3647_WELDING.pdf
• https://www.cdc.gov/niosh/docs/88-110/pdfs/88-110.pdf
• https://www.cdc.gov/niosh/topics/welding
• https://www.osha.gov/chemicaldata/501
• https://stacks.cdc.gov/view/cdc/196846/cdc_196846_DS1.pdf
• https://www.assemblymag.com/articles/86511-troubleshooting-the-mig-welding-process
• https://unimig.com.au/blog/troubleshooting-your-weld-the-12-most-common-problems-and-how-to-fix-them