Crappy Welds: Why They Happen, How to Diagnose Them, and How to Fix Every Common Defect

What Actually Makes a Weld “Crappy”? Defining Weld Quality

You flip up your hood, and there it is. A lumpy, spattered, pitted bead sitting on the plate like chewed bubble gum. You know it’s bad. What you’re not sure about is why it happened or what to do next. That gap between spotting a bad weld and understanding it is where real skill gets built.

So let’s start with the obvious question: what actually makes a weld “crappy”? The honest answer is that “crappy” isn’t a technical term at all. It’s a gut feeling, and gut feelings have to be translated into something a code, a customer, or an inspector can actually act on.

Discontinuity vs. Defect: A Crucial Distinction

“Crappy” is a lay term. Welding codes use two more precise words. A discontinuity is any interruption in the weld’s intended structure, like a pocket of gas, a crack, or a wavy edge. A discontinuity only becomes a defect when it exceeds the acceptance limits of the applicable code.

That last part is easy to miss. The same porosity might be perfectly acceptable on a backyard gate and a rejection on a bridge. Codes such as AWS D1.1 define visual acceptance criteria for undercut, porosity, and weld profile, so “good enough” is always defined by the standard you’re working to. Elongated piping porosity is a good example: because it’s typically confined to the face reinforcement of a groove weld or the convex face of a fillet weld, AWS guidance treats it as a workmanship issue rather than a structural one.

The Three Pillars of a Sound Weld

Strip away the jargon and every acceptable weld has to nail a few fundamentals:

• Fusion — the filler and base metal must actually melt and merge into one continuous material, not just sit next to each other.
• Penetration — the weld must reach deep enough into the joint to carry the load it’s designed for.
• Consistency — the bead should hold steady size, shape, and quality along its whole length. A joint is only as strong as its weakest point.

Cracks get their own category because they’re almost never tolerable. A crack concentrates stress and spreads, which is why most codes reject them outright regardless of size. Miss one on a loaded part and you’ve built in a failure that will grow on its own.

Appearance Is a Clue, Not the Whole Story

Here’s the trap beginners fall into. A weld that looks rough is often weak, but a pretty, uniform bead can still hide lack-of-fusion or a buried crack. Appearance tells you something went wrong; it doesn’t prove a weld is sound. This is why reading a bead properly means correlating what you see with what your settings and technique are actually doing. The surface is the first diagnostic tool, not the final verdict.

How This Guide Will Help

Over the next sections we’ll turn that vague frustration into a repeatable diagnosis. First we’ll break down the major defect families behind most bad welds, then show you how to read a bead like an inspector. From there we’ll trace root causes in both settings and technique, walk through process-specific problems in MIG, flux core, stick, and TIG, and cover when to grind a weld out versus redo it. We’ll finish with prevention, the safety risks that bad welds and welding itself can create, and a practical roadmap for improving pass by pass — the kind of deliberate work structured welding practice is built around. By the end, a crappy weld won’t just be something you notice. It’ll be something you can explain, fix, and stop making.

The Big Five Defect Families Behind Most Crappy Welds

Almost every bad bead you’ll ever pull traces back to one of five defect families. Learn to recognize them on sight and you’ve already solved half the diagnostic puzzle. Each one damages a weld differently, and each one points to a different root cause.

Porosity: Trapped Gas in the Puddle

Porosity is the Swiss-cheese problem. Gas gets trapped as the weld pool solidifies and leaves rounded voids, either scattered across the surface or buried inside the bead. The usual culprits are contamination like oil, paint, rust, or moisture; damp electrodes or wire; or shielding gas that never fully covered the pool. A gusty fan, a clogged nozzle, or a worn contact tip can all pull air into the arc.

Porosity is a volumetric defect — a three-dimensional cavity inside the metal. A few isolated holes are mostly cosmetic, but clusters or continuous porosity cut the load-bearing cross-section and signal that your shielding failed. Check gas flow and consumables first.

Lack of Fusion and Cold Lap

This is the quiet killer. Lack of fusion means the weld metal never completely melted into the base metal, or into the preceding bead. On the surface it can look disturbingly fine, but underneath the two metals are merely sitting next to each other. Cold lap is the same failure in fillet welds, where the bead rolls over the toe without wetting in. Both come from too little heat, travel speed that outruns the puddle, or an arc aimed at the wrong spot.

Because a lack-of-fusion boundary is a flat, unbonded plane, it behaves like a crack waiting to open. That makes it one of the more dangerous defects you can hide in a joint.

Undercut: The Sharp Notch

Undercut is a groove melted into the base metal along the weld toe and left unfilled. Too much current, too long an arc, or the wrong travel angle will do it. What makes undercut nasty is its shape: a sharp notch running along the edge of the joint, concentrating stress. Under repeated or fatigue loading, that notch is exactly where a crack starts. The AWS D1.5 bridge welding code even tightened its undercut tolerances in 2025, a reminder of how seriously the notch is taken.

Cracking: No Wiggle Room

Cracks are the one defect with no acceptable excuse. Hot cracks form as the weld solidifies; cold cracks appear hours or days later, often driven by hydrogen or restraint. In most structural codes a crack is never permitted. You cannot weld over it and call it fixed, because the crack tip keeps propagating under load. It has to be ground out to sound metal and rewelded.

Slag Inclusions and Overlap

Slag inclusions show up in flux-based processes like SMAW and FCAW when flux fails to melt out and rise to the surface, so it stays trapped between passes. Like porosity, it’s a volumetric defect: it weakens the weld and gives corrosion a place to start. Overlap is different — weld metal that spills past the toe and rolls onto cold base metal without fusing.

Volumetric vs. Planar: Why Shape Matters

Now the distinction that separates a cosmetic flaw from a structural one.

• Volumetric defects (porosity, slag): Rounded, three-dimensional cavities. They reduce cross-section and look ugly, but their smooth shapes don’t concentrate stress sharply.
• Planar defects (cracks, lack of fusion, undercut): Flat faces or sharp notches. These act as stress risers, and under fatigue loading they are the ones that fail, because stress piles up at the sharp tip.

That’s why a small crack can be far more serious than a cluster of porosity. Some defects are merely cosmetic; others quietly steal strength or hand a crack its starting point. For a visual reference on what healthy metal should look like, see our guide to good and bad MIG welds. The ESAB welding defects guide and Miller’s troubleshooting resources both break these down further.

Learn these five families and you’ll never look at a crappy weld the same way again. Next, we’ll read the bead itself.

Reading the Bead: How Weld Appearance Reveals the Problem

You just finished a pass and something looks off. Before you reach for the grinder, look closer. The bead is talking to you, and with a little practice you can read it as clearly as a gauge. Most crappy welds announce their cause right on the surface, if you know which features to check.

Bead Profile: The Shape Tells the Temperature

The cross-section of your bead is the fastest clue to heat input. A convex, bulging bead that looks like a rope laid on top of the plate indicates the weld is running too cold. The metal piles up instead of wetting out, so the toes never fuse properly. The fix is more heat or an angle adjustment that pushes that heat into the joint, not the puddle’s center.

A concave or sunken bead is the opposite story: too much heat, or you’re traveling too fast for the settings. The puddle sags below the surface, and on thin material that same excessive heat input causes warping, distortion, or outright burn-through.

Two more shapes are worth memorizing. A narrow, ropy bead usually means travel is too fast or voltage is too low; because the arc doesn’t linger, you get inadequate tie-in at the toes and insufficient penetration. A wide, flat, washed-out bead points to too much heat for the material thickness.

Spatter, Arc Wander, and Wobbly Widths

Spatter is a symptom, not a random annoyance. Excessively high voltage, travel that’s too fast, and excessive wire stickout all throw molten metal across the work. As a rule, your wire stickout should not exceed about 1/2 inch past the nozzle. Pulling the gun farther back changes the arc length and invites spatter and cold lap.

Spatter tells its own story: a fine, even spray usually means you’re close to dialed in, while coarse, heavy globs flung well away from the joint suggest the arc is too long or the voltage is set high for your wire and gas.

If the bead weaves side to side on its own, that’s arc wander, and it often traces back to wire-feed trouble, a worn contact tip, or dirty base metal. Inconsistent bead width, where the pass is fat in one spot and skinny in the next, almost always means your travel speed varied. That’s a rhythm problem, not a settings problem.

Color Still Tells the Truth on Stainless

On stainless steel, the heat tint (that “rainbow” from straw to gold to blue to gray) is a map of heat input and shielding quality. A light straw or gold color indicates good shielding and controlled heat. As the tint moves toward blue and dull gray, heat input or shielding breakdown is rising, and you’re cooking out corrosion resistance. Getting this right starts with polarity and gas coverage, which is why proper polarity matters so much for precision work.

A Quick Visual Diagnostic Table

Watch the toes, not just the crown: a bead can look reasonably smooth on top while the edges fail to blend into the plate, and that lack of fusion is what actually fails under load. Once you can name the defect on sight, the repair gets far more targeted. For a deeper look at the classic surface signals, see our guide to good and bad MIG welds, then turn the fixes into habits with structured welding practice.

Root Causes Part 1: Settings, Machine Setup, and Consumables

Before you blame your hands, blame your machine. A huge share of crappy welds trace straight back to the knobs, the leads, and the consumables — the equipment side of the equation that too many welders set once and forget. When a bead goes bad, the settings and setup are the fastest things to rule out, and often the easiest to fix.

Voltage and Amperage Mismatch

Voltage controls arc length; amperage controls heat and deposition. On a wire welder, wire feed speed is effectively your amperage. Get the two out of sync and the weld tells you immediately. Too much voltage for a given wire feed speed gives you a long, wandering arc that spatters and spreads wide. Too little voltage chokes the arc and stubs the wire into the puddle.

The same logic applies to stick and TIG amps. Too cold and you get poor fusion and a lumpy, sitting-high bead; too hot and you blow holes or undercut the edges. Match the current to thickness, joint, and rod or wire size. Most machines publish a door chart with recommended voltage and wire feed speed ranges, so start there and fine-tune from the bead.

Wire Feed Speed and Stickout

Feed wire too slowly and the arc rides high and cold; feed it too fast and the wire rams into the work before it can melt, giving you a stuttering, spattery mess. Then there’s stickout. Extending the wire too far from the nozzle, beyond about 1/2 inch, invites porosity and spatter and throws off your heat because the wire preheats differently. Keep stickout short and consistent. A slipping drive roll or too little spool tension mimics bad settings, so check the feed mechanism before you ever touch a dial.

Wrong Polarity

Polarity decides which direction the heat flows. Self-shielded flux-cored wires should be run with straight polarity (electrode negative), while most gas-shielded wires want reverse polarity (DCEP). Wire it backwards and you get a cold, sputtering, spatter-heavy bead that no amount of technique can rescue. Brush up on polarity for flux core welding before you spool up. Polarity matters on TIG and DC stick too, because it shifts where the heat concentrates and changes penetration.

Gas Flow, Gas Quality, and Wind

Porosity in MIG is most often caused by poor shielding: restricted, turbulent, or insufficient gas flow, or wind displacing the gas before it can protect the puddle. Typical shielding gas flow for MIG is roughly 30–40 cubic feet per hour. Run too low and you leave the puddle exposed; crank it too high and the turbulent stream actually pulls air into the arc. Damp or contaminated gas does the same damage, so inspect your regulator, hoses, and cylinder. Check for leaks at the fittings as well; a slow leak bleeds off flow without you touching the dial. Miller’s weld defect troubleshooting guide confirms how often gas issues sit behind porosity complaints.

Worn Contact Tips, Liners, and Undersized Guns

Consumables wear out, and worn parts quietly sabotage good settings. A worn or oversized contact tip lets the wire wander, so the arc scatters and spatter climbs. A dirty, kinked, or wrong-size liner drags on the wire and causes erratic feed that looks exactly like a machine fault. And an undersized MIG gun pushed past its duty cycle overheats, delivering unstable current and a ragged bead. Match the gun to the job and replace tips and liners on schedule.

Ground Clamps, Leads, and Arc Blow

A poor-quality or poorly placed ground clamp can cause erratic arcs and inconsistent weld results. Clamp to clean, bare metal close to the weld, not to a painted or rusty surface. Loose, corroded, or undersized quick connect welding leads add resistance and heat, starving the arc of stable current. A clean magnetic or C-clamp ground on bare steel keeps current steadier than a weak spring clamp on rust. Add arc blow — the magnetic deflection of the arc, common in DC welding of thick or corner sections — and the puddle drifts off course no matter how steady your hand is. Lincoln Electric’s MIG problems and remedies guide covers these current and feed faults in useful depth.

Root Causes Part 2: Technique, Travel, and Human Factors

Your machine can be dialed in perfectly and your consumables fresh out of the box, and you’ll still lay a crappy weld if your hands don’t do the right thing. Technique is where most frustrating beads are born. The good news? Every one of these problems has a physical cause you can see and correct.

Travel Speed and Arc Length: The Two Variables You Feel

Travel speed controls how much heat and filler land in a given length of joint. Move too slow and you pile up a wide, convex bead with excessive heat input. Too fast and the puddle never fully wets the sidewalls, leaving a narrow, ropey bead with cold lap at the edges. Arc length (or stickout) is just as important. Stretch the arc too long on stick or TIG and you lose penetration, spatter, and shielding. Bury the electrode too deep, particularly with MIG, and you risk the wire stubbing into the puddle. On wire processes, stickout is a real tuning tool, which is why contact tip wear and stickout change your results even when the dials never move.

The fix is boring but effective: watch the puddle, not the arc. If the puddle is the right size and the toes are wetting in, your speed is right. You can confirm the result against photos of good and bad MIG welds so your eye learns what “right” actually looks like.

Travel Angle and Work Angle: The Undercut Generator

These two get confused constantly. Travel angle is the tilt along the direction of travel (push or drag). Work angle is the tilt across the joint, side to side.

Undercut is the classic work-angle defect. You can create it when you focus the arc too much on one plate, which happens naturally when your work angle leans toward that side and melts the edge without filling it back in. Excessive voltage combined with too-fast travel will do the same thing. Adjusting your welding angle by just 5 to 10 degrees can significantly improve penetration and bead shape, because that small change redirects heat to the joint root where it belongs.

Riding the Pool: The Silent Lack of Fusion

“Riding the pool” means letting the arc ride on top of the molten metal instead of driving into the joint. The puddle looks great, the bead looks fine, and underneath there’s no fusion between the filler and the base metal. Welding current that is too low is the most common cause of lack of fusion, but so is a lazy travel angle that keeps the arc ahead of the puddle without digging in. The remedy is to aim the arc at the leading edge of the puddle and push metal into the root rather than layering it on top.

Crater Cracks and Stopping Short

When you snap the arc off before the crater is filled, you leave a shallow, unfilled dish of molten metal that shrinks and tears as it cools. Crater cracks form when the end of a weld isn’t filled properly, creating a weak point that cracks. The cure is simple: pause, add extra filler, and either backstep the arc or use a crater-fill function. Never just pull the trigger and walk away.

Weave Technique: Pause at the Toes

Weaving isn’t about waving the gun around for looks. It’s about managing heat and filling the joint evenly. The key move is pausing longer at each side of the bead when weaving, which helps wash filler metal into the toes and can reduce undercut. Move across relatively quickly, then hesitate at each toe long enough to let the puddle catch up.

Body Position, Fatigue, and Consistency

Ignoring ergonomics is a strategy for crappy welds. If you’re straining, reaching, or fighting your posture, your travel speed and angle will drift within a single bead. Brace your hands or forearms against the work, get comfortable before you strike an arc, and break up long sessions. Fatigue shows up as inconsistency long before you feel tired.

And here’s the honest truth: none of this clicks from reading alone. It clicks from repetition. If you want to build the muscle memory that makes angle and travel speed automatic, work through our welding practice guide and treat every scrap bead as a data point.

Process-Specific Crappy Welds: MIG, Flux Core, Stick, and TIG

The same ugly bead can point to completely different causes depending on which process laid it down. A defect that screams “bad gas” on MIG means nothing on stick, and the fix for a TIG problem will do nothing for flux core. Here’s how each process fails, and what to do about it.

MIG (GMAW): Gas Trouble and Cold Lap

MIG lives and dies by its shielding gas, so porosity is the signature defect. A weak or disrupted gas flow, a clogged nozzle, a breeze across the bench, or too much stickout all let atmosphere into the pool and leave those pinhole craters behind. Check your flow rate at the nozzle, not the regulator, and clean spatter out of the gun regularly. Contact tip wear matters more than most beginners realize, since a loose or egg-shaped tip wrecks wire feed and arc stability, so keep spares on hand per this contact tip troubleshooting guide.

Cold lap (lack of fusion) shows up as a bead that sits on top of the metal instead of melting into it. It usually means your voltage is too low, travel speed is too fast, or you’re aiming at the joint’s center instead of its leading edge. On thin sheet, the opposite problem strikes: burn-through, caused by too much heat and not enough speed.

Aluminum deserves its own warning. It builds up an oxide layer that must be removed before welding, and you should use a stainless steel brush to avoid contaminating the weld. Feeding aluminum wire has been likened to pushing a wet noodle through a straw, which is why birdnesting, where the wire tangles at the drive rolls, is so common and costly. Loosen the tension and switch to a spool gun or push-pull setup.

Flux Core (FCAW): Slag and Polarity Traps

Gasless flux-cored welding relies on a flux core rather than an external shielding gas. It’s more forgiving of wind, but that same flux creates slag, and slag inclusion happens when flux falls off and ends up trapped in the weld pool. Clean between every pass, and don’t drag your nozzle through hot slag. If you’re unsure how this process differs from MIG, see is flux core the same as MIG.

Polarity mistakes are a huge source of flux-core trouble. Most gasless wires run DCEP (electrode positive), and hooking up the leads backwards produces a spattery, cold, unstable arc. Get it right with a dedicated polarity for flux core welding guide, and if you’re going the gasless route for the first time, the gasless MIG welding guide covers setup in depth.

Stick (SMAW): Slag Entrapment, Arc Blow, and Restarts

Stick brings its own trio of headaches. Slag entrapment occurs when you fail to chip and brush thoroughly between passes, or when your arc angle drags slag into the pool. Arc blow, the wandering of the arc caused by magnetic fields, shows up especially near the ends of a joint or on heavy sections, and it produces erratic, spattery beads with poor penetration. Shorten your arc, reduce current, change your work angle, or move your ground clamp closer. Poor restarts are the last common gripe: the crater and cold start that follow every rod change. Strike ahead of the crater and back into it to avoid a cold lap. Proper amperage helps too, as covered in this E7018 amperage and technique breakdown.

TIG (GTAW): Cleanliness and Heat Control

TIG punishes sloppiness. Dip the tungsten and you get tungsten contamination, a sputtering arc and a contaminated pool, so stop and regrind immediately. Arc wander usually comes from a contaminated or badly shaped electrode. Poor heat control causes either a cold, unfused joint or a burned-through mess. Since TIG’s heat balance depends on polarity, understanding TIG polarity: DCEN, DCEP, and AC is essential, especially for aluminum.

Whatever the process, prevention follows the same logic: match settings to the metal, keep consumables and surfaces clean, and practice deliberately. Building that muscle memory is exactly what our guide to welding practice walks you through.

Repairing Crappy Welds: When to Grind, When to Reweld

Before you reach for the grinder, answer one question: is the defect cosmetic or structural? Cosmetic issues—slight spatter, uneven ripples, minor surface roughness—affect appearance, not load capacity. Structural defects—cracks, lack of fusion, deep undercut, excessive porosity—compromise the joint itself. Cosmetic flaws usually need only a dressing pass. Structural flaws demand a real repair, and guessing wrong can turn a small problem into a failed weld. Our breakdown of good and bad MIG welds helps you tell them apart at a glance.

The Repair Decision Tree

Work through it in order:

1. Identify the defect — its depth, length, and location.
2. Classify it as cosmetic or structural.
3. Choose the method: blend, grind out, or remove and reweld.
4. Re-inspect before calling it done.

Every good repair starts with a clean surface. Grinding, wire brushing, and solvent wiping remove the contamination that caused many defects in the first place.

Undercut: Grind or Reweld?

Undercut is a groove melted into the base metal along the weld toe, and it concentrates stress right where the joint is most vulnerable. The 2025 AASHTO/AWS D1.5 Bridge Welding Code introduced a preference for addressing undercut by grinding and blending it out rather than adding more weld metal, provided the undercut is not too deep. Grinding must leave a smooth surface with a gentle transition. If the resulting depression exceeds 1/32 inch, the undercut must be repaired by welding instead.

Porosity

Porosity is gas trapped in the bead. Scattered, tiny pores within limits may pass visual inspection—AWS D1.1 includes visual acceptance criteria for piping porosity—but they still mark weak spots. For repair, grind the porous area back to sound metal and reweld. Skimming filler over the top only traps the gas again, and porosity that exceeds acceptance criteria must be removed, not blended.

Cracks: Excavate, Never Cap

Cracks are the most serious defect you’ll face. You cannot just go over the top of a crack and fill it in—the cracked portion, or the entire weld, must be removed and restarted. Excavate with a grinder or arc gouging until you reach sound metal, then reweld. Confirm the excavation with dye penetrant before you strike an arc, so you don’t bury a crack tip under fresh metal.

Incomplete joint penetration in a double groove weld is often caused by insufficient back gouging or inadequate access geometry. Correct the cause—improve access, back gouge properly—before rewelding, or the defect returns.

Overlap and Excess Reinforcement

Overlap is weld metal that rolls over the toe without fusing—a structural defect, not a cosmetic one. Grind it back to a proper toe. Excess reinforcement (a convex, lumpy cap) creates a stress riser, so blend it down to size. Our guide to the overlap welding defect covers detection and repair in detail. Never feather a flaw into looking acceptable—grind until the profile matches the surrounding metal.

Post-Repair Inspection

A repaired weld is a new weld and deserves fresh inspection. Clean it, check dimensions and profile, and repeat the visual checks you’d apply to any first pass. Structural repairs may require dye penetrant, magnetic particle, or ultrasonic testing. Document what you did—defect type, method, and result—and re-test the joint under load where the application demands it.

When a defect is deep, widespread, or keeps returning, don’t patch—cut the section out and reweld from scratch. A clean restart beats a stack of band-aids every time, and skill here is built the same way as anywhere else: deliberate repetition and honest self-review, which is what our welding practice guide is all about.

Preventing Crappy Welds: Prep, Fit-Up, Settings, and Practice

Every defect has a cause, and almost every cause is something you control before the arc ever strikes. Prevention is not glamorous, but it is the cheapest skill you will ever build. Nail these four habits and the lumpy, spattered beads start disappearing from your work.

Prep the Metal Like It Matters

Surface contaminants such as oil, rust, paint, or grease can prevent proper weld penetration and lead to porosity. Grind, wire-brush, or solvent-wipe the joint down to bright, bare metal before you fit anything up. This matters most with GMAW, which tolerates far less surface contamination than slag-generating processes. The slag blanket on stick or self-shielded flux-core protects the puddle from some dirt; MIG has no such backup, which is why manufacturers stress clean base metal in their troubleshooting guides. For aluminum, remove the oxide layer right before welding and use a dedicated stainless steel brush so you are not smearing contamination back in.

Fit-Up, Clamping, and Tacking

A bad joint cannot be welded into a good one. Set your gap, alignment, and bevel to the drawing, then clamp the parts so they stay put as heat tries to pull them out of position. Tack in a balanced sequence, working opposite ends and staggered points, to lock geometry without fighting distortion later. Finally, clamp your ground on clean bare metal close to the weld. A solid electrical ground with a firmly connected clamp near the joint improves arc consistency and eliminates a whole category of wandering, sputtering “why is this thing fighting me” welds.

Dial In Settings Before You Commit

Parameter selection starts with material thickness and process, not with whatever the last person left the dial on. Too little amperage or wire speed gives cold lap and poor fusion; too much burns through thin stock and invites undercut. Polarity matters just as much, so confirm the correct setup for your process, whether that means reviewing TIG welding polarity, checking polarity for flux core welding, or finding a reliable 3/16 E7018 amperage starting point.

Keep Consumables and Equipment Honest

Worn consumables quietly sabotage good technique. A mushroomed or oversized contact tip lets the wire wander, a clogged liner causes erratic feed, and a spattered nozzle disturbs gas coverage. Build a simple schedule: inspect the tip daily, replace it when the bore wears oval, blow out the liner, and check gas flow and hose connections. Treating torch parts as wear items rather than permanent fixtures prevents a surprising share of mystery defects.

Practice With Purpose

Recommended practice includes practicing on scrap material before tackling difficult jobs, but the quality of that practice matters more than the hours. Running bead after bead while tired just grooves in your mistakes. Instead, run deliberate sessions: pick one variable, like travel angle, arc length, or speed, and change nothing else. Weld sample coupons, cut them, and inspect the cross-section for fusion and penetration. That is how welding practice builds real skill instead of muscle memory for bad habits. New to a wire welder? Learn how to use a wire welder properly before you chase settings.

A Pre-Weld Checklist You Can Actually Use

• Clean the joint to bright metal: remove rust, mill scale, oil, paint, and moisture.
• Degrease aluminum and brush off the oxide with a stainless brush.
• Check fit-up, gap, and bevel; clamp and tack in a balanced sequence.
• Clamp the ground firmly to clean bare metal close to the weld.
• Confirm process, polarity, amperage or voltage, and filler match the thickness.
• Inspect the contact tip, nozzle, liner, and gas flow.
• Run a test bead on matching scrap before touching the real part.

A planned workspace underpins all of it: good lighting, a stable bench, ventilation, and organized tools turn good habits into default behavior, which is what a well-planned welding facility setup delivers.

The Hidden Danger: Why Bad Welds and Welding Itself Can Hurt You

A crappy weld is easy to laugh off when it’s a practice coupon. On a trailer hitch, a roll cage, or a bridge, it’s a different story. Voids, incomplete fusion, undercut, and overlap act as stress concentrators. Under repeated loading they turn into fatigue cracks that can grow silently until the joint fails, sometimes with no visible warning. Codes exist because defect tolerances have real consequences; the 2025 AWS D1.5 bridge welding code tightened undercut limits for exactly this reason.

Undercut, overlap, and incomplete fusion are not cosmetic defects. Never trust a load-bearing joint unless it’s been made and inspected properly. If you’re unsure whether a defect is acceptable, treat it as structural until proven otherwise.

Fatigue is the sneaky part. A weld that passes a quick visual check can still harbor incomplete fusion at the root, and cyclic stress finds it. That’s why AWS publishes qualification standards such as B2.1, which tie procedures and welder performance to documented testing rather than appearance.

The Air You Breathe Is the Daily Hazard

Structural failure is dramatic and rare. Fume exposure is the quiet risk that shows up on every pass. OSHA sets a permissible exposure limit for welding fume of 5 mg/m³ as an 8-hour time-weighted average, measured in the welder’s breathing zone. Acute exposure to welding fume and gases can cause eye, nose, and throat irritation, dizziness, and nausea. Prolonged exposure may cause lung damage and cancers including lung, larynx, and urinary tract cancers. The NIOSH welding safety overview covers these effects in more detail.

Chromium in stainless steel converts to the hexavalent state, Cr(VI), during welding. OSHA regulates it under 29 CFR 1910.1026 and 1926.1126 with a PEL of 5 µg/m³ as an 8-hour TWA, roughly a thousand times stricter than general fume. Manganese is another concern, linked to neurological effects in overexposed welders. And zinc coatings bring metal fume fever: flu-like chills, fever, and a metallic taste. It has been estimated that 30 to 40 percent of all welders have experienced metal fume fever at some time.

Arc Flash, UV, and Personal Protection

The arc also attacks you directly. Ultraviolet and infrared radiation cause “arc eye,” skin burns, and long-term eye damage, so a properly shaded helmet is non-negotiable. See our guide to protective eyewear for welding. Wear safety glasses under the hood for chipping and grinding, plus leather gloves, flame-resistant sleeves, and a respirator when fume is present.

Ventilation: Engineering Out the Exposure

PPE is the last line of defense; ventilation is the first. Local exhaust ventilation (LEV) captures fume at the arc before it reaches your breathing zone. Fume extraction guns, downdraft tables, and hooded arms all count. In enclosed spaces, general dilution ventilation plus supplied-air respirators may be required. Setting this up correctly is a shop-design decision, and our guide to welding facilities walks through layout, airflow, and safe work zones.

Why Cleaner Technique Is a Safety Control

Here’s the hopeful part: safety and skill overlap. Slag, spatter, and porosity mean more grinding, more chipping, and more airborne dust, so every defect you eliminate cuts your exposure. Running correct parameters, keeping your contact tip in good condition, and dialing in travel speed all reduce spatter and fume at the source. That’s why deliberate, structured welding practice pays off twice: better beads and a healthier pair of lungs.

From Crappy to Confident: A Practical Improvement Roadmap

Every welder who has ever struck an arc has produced a crappy weld. That lumpy, spattered bead isn’t proof you lack talent — it’s data. The welders who improve fast aren’t the ones born with steady hands; they’re the ones who treat each bad weld as a clue instead of a verdict.

The Diagnostic Loop: Five Steps, Every Time

Turn every rough pass into a repeatable routine:

1. Observe the bead up close under good light.
2. Identify the defect — porosity, undercut, overlap, cold lap, spatter.
3. Trace the likely cause using what the bead is telling you.
4. Adjust exactly one variable — voltage, travel speed, angle, wire feed, gas flow.
5. Re-test on scrap and compare the new bead against the old one.

Changing only one variable at a time is the whole trick. Tweak three settings at once and you’ll never know which fix actually worked. This is where bead reading pays off: learning to interpret good and bad MIG welds speeds your improvement more than any single tip you’ll ever read.

Keep a Weld Log

Write down the material, thickness, process, polarity, amperage, voltage, wire or rod, gas flow, and your honest result. Then connect symptoms to causes. Spot elongated porosity along a bead edge and that often points to an electrode extension that’s too short — lengthen it slightly, amperage drops, and the problem frequently disappears. A worn contact tip or a clogged liner can mimic that same ugly result, so note those too. Ten seconds of logging saves an hour of guessing.

Consistency Beats Occasional Brilliance

The welder who lays a decent bead every single time will always out-earn the one who nails a perfect pass once a month. That reliability is built, not born. Consistent practice on scrap coupons builds the muscle memory that prevents most beginner-level crappy welds: the steady travel speed, the disciplined stickout, the confident drag angle. As Miller’s defect troubleshooting guidance stresses, most failures trace back to setup and technique you can control and train.

So keep practicing, and inspect your own work with a critical eye — nobody else will catch your weak pass before a code inspector does. Suit up properly with welding safety gear that protects your eyes, then burn through another coupon. Crappy welds aren’t a dead end. They’re the raw material every good welder was built from.

Frequently Asked Questions

What is the most common cause of a crappy-looking weld?

For MIG, poor shielding gas coverage — from low flow, leaks, drafts, or too much stickout — is the leading cause of porosity and ugly beads. For most processes, incorrect heat settings and travel speed combinations come next, producing convex, narrow, or spattered beads.

Can a bad-looking weld still be strong?

Sometimes yes. A slightly rough or convex bead may still have full fusion and penetra­tion, just poor appearance. But looks are a warning sign: undercut, cracks, and lack of fusion are structural defects you cannot safely ignore.

How do I fix undercut on a weld?

Shallow undercut within code limits is often repaired by grinding it to blend smoothly into the base metal. If the resulting depression or the original undercut exceeds allowable depth (commonly around 1/32 inch), it must be repaired by welding.

Why does my MIG weld keep getting porosity?

Check shielding gas first: verify flow (about 30–40 CFH), look for hose and gun leaks, block drafts, and keep wire stickout under about 1/2 inch. Then clean the base metal — oil, rust, paint, or moisture all generate porosity.

Is it safe to just weld over a crack?

No. A crack must be fully excavated and removed before rewelding, because simply capping over it traps the defect and it will reopen. Grind out the cracked metal completely, then weld and re-inspect.

Does polarity matter for avoiding crappy welds?

Absolutely. Self-shielded flux-cored wire generally runs on electrode negative (DCEN/straight polarity), while many MIG setups use electrode positive (DCEP). Using the wrong polarity causes instability, spatter, and poor fusion.

How can I stop producing bad welds as a beginner?

Clean your metal, set parameters from the machine chart and then fine-tune to the bead, practice on scrap coupons, and change only one variable at a time. Consistent, deliberate practice builds the muscle memory that prevents most beginner defects.

Are welding fumes from bad welds dangerous?

Welding fume is hazardous regardless of weld quality, but excess spatter and grinding add particulates. Use local exhaust ventilation, avoid welding in confined spaces without ventilation, and wear respiratory protection when controls are insufficient.

Authoritative Sources & Further Reading

• https://www.osha.gov/sites/default/files/publications/OSHA_FS-3647_WELDING.pdf
• https://www.cdc.gov/niosh/npg/npgd0666.html
• https://www.cdc.gov/niosh/welding/about/index.html
• https://www.cdc.gov/niosh/chemicals/pel88/pell-pages/welding.html
• https://www.cdc.gov/niosh/docs/88-110/pdfs/88-110.pdf
• https://www.aws.org/magazines-and-media/welding-digest/2025/july/wd-july-2025–new-undercut-tolerances-and-treatments-in-the-2025-d15-bridge-welding-code
• https://www.aws.org/magazines-and-media/inspection-trends/it-aug-25-the-answer-is
• https://pubs.aws.org/Download_PDFS/B2.1-B2.1M-2021-Web_PV.pdf
• https://www.millerwelds.com/en-us/resources/knowledge-hub/mig-welding/how-to/troubleshooting-weld-defects
• https://www.lincolnelectric.com/en/welding-and-cutting-resource-center/process-and-theory/mig-problems-and-remedies
• https://www.hobartbrothers.com/resources/technical-articles/mig-troubleshooting-for-metal-cored-and-solid-wire
• https://esab.com/al/eur_en/esab-university/articles/welding-defects-guide