Good Aluminum Welds: How to Recognize, Produce, and Inspect Them

Table of Contents

What Makes a Good Aluminum Weld? Defining Quality Beyond Looks

Run a clean bead down a T-joint, lift your hood, and the weld looks like a textbook illustration. Then you tap it with a chipping hammer and a hairline crack opens right down the centerline. Aluminum hides its problems beautifully, and that gap between appearance and reality is exactly why “good” needs a definition. A good aluminum weld is not the one that photographs best. It is the one that satisfies four tests at the same time.

The Four Tests of a Good Aluminum Weld

Judge every weld against these four, in order.

Sound fusion. The filler and base metal must melt and merge completely at the root and sidewalls. No lack of fusion, no incomplete penetration, no cold laps that look fine on the surface and carry nothing.
Freedom from defects. That means no porosity, no cracking (hot or cold), and no inclusions of oxide, tungsten, or slag trapped in the deposit.
Correct profile and dimensions. Leg size, throat, bead height, and toe angles have to match the drawing. A weld can be perfectly sound and still be wrong if it is undersized or badly shaped.
Adequate mechanical properties. The joint, and the heat-affected zone around it, must meet the strength and ductility the application demands.

These four are interdependent. A sound-looking weld with hidden porosity still fails the mechanical test, and a weld with perfect penetration is scrap if it’s undersized. Miss any one and the joint is defective, no matter how good it looks.

Why a Pretty Bead Can Still Be a Bad Weld

Most aluminum defects are invisible from the outside. Porosity sits below the surface where the arc couldn’t let trapped gas escape. Centerline cracking can hide inside a smooth, rippled cap. Lack of fusion along a sidewall is often covered by a full-looking bead, and aluminum’s high thermal conductivity pulls heat away fast, making those cold laps even easier to create. Because aluminum oxide melts at roughly 3,700°F (2,055°C) while the metal beneath it melts near 1,220°F (660°C), a weld can look wet-in while the oxide film kept the two sides from ever truly joining. Appearance tells you technique. It cannot tell you what happened underneath.

A Code-Aware Framework, Not Just a Look Test

For structural aluminum, the governing reference is AWS D1.2/D1.2M:2026, Structural Welding Code—Aluminum, the first update since 2014. Two things surprise people coming from steel. First, unlike the steel code, D1.2 does not allow prequalified procedures; every joint detail and welding procedure must be individually qualified in a Welding Procedure Specification (WPS). That makes procedure qualification, not operator confidence, the starting point of quality. Second, you cannot assume the weld matches the parent metal. As-welded joint strength in aluminum is rarely equal to the base material, and HAZ strength loss is often 30% or more, which is why heat input and joint design matter as much as bead shape. Acceptance is judged by visual, radiographic, and dye-penetrant testing methods per D1.2.

Quick Note: In the U.S., structural aluminum welding falls under AWS D1.2. For the general industry safety rules around welding fumes, arc radiation, and PPE, OSHA’s welding and cutting page is the baseline reference.

What This Guide Delivers

So “good” is a checklist, not a feeling: fusion, freedom from defects, correct dimensions, and real mechanical properties, all judged against a qualified WPS. The rest of this guide works through that checklist in order. We’ll cover the metallurgy that makes aluminum behave so differently from steel, how to read a bead and spot trouble, prep and filler selection, settings for AC TIG and MIG, the defects that ruin welds, inspection methods, and the field-tested fixes that separate consistent welders from lucky ones. If you’re coming from steel, our guide to good and bad MIG welds makes a useful side-by-side comparison, since the same bead that signals a healthy steel weld can mean something very different on aluminum.

Why Aluminum Is Different: The Metallurgy Behind Good Welds

If steel is the dependable friend who always shows up on time, aluminum is the brilliant, moody cousin. It welds differently because it is different at the atomic level, and every technique you’ll read about in this article flows from the material properties below. Skip the science and you’ll spend years fighting defects you don’t understand. Understand it and the settings, prep, and torch angles start to make obvious sense.

The Oxide Layer That Refuses to Melt

Aluminum reacts with air almost instantly, forming a thin, tough skin of aluminum oxide (Al₂O₃). The trouble is that the melting points are wildly mismatched. Aluminum melts at about 660 °C (1220 °F), while its oxide melts at roughly 2000–2050 °C (about 3762 °F). In practical terms, that film stays solid while the metal underneath is already a liquid puddle. You’re trying to weld through a ceramic crust sitting on top of water.

That oxide also happens to be an electrical insulator. It can block arc initiation entirely or cause erratic commutation (unstable current transfer) inside a MIG contact tube, which is one reason worn or incorrect consumables cause so much grief. This is exactly why contact tip welding maintenance matters more on aluminum than steel.

The fix is a combination of mechanical or chemical cleaning and the right arc action. Alternating current in TIG spends part of each cycle cleaning the surface, which is why TIG welding polarity is such a big deal for aluminum and why specialized approaches to DC welding aluminium exist as an alternative. But the cleaning doesn’t last: the oxide begins to reform immediately after you scrub it away, often within one to two hours. Prep right before you weld, not the night before.

Heat That Runs Away From You

Aluminum’s thermal conductivity is roughly three times that of steel, and it conducts heat almost as well as copper. On the one hand, that’s a blessing: heat spreads out and the part cools quickly. On the other, it means you need more amperage and a hotter start just to get the puddle going, because the surrounding metal is stealing your heat as fast as you add it.

The high electrical conductivity compounds things. Aluminum carries current so easily that it doesn’t build up resistance the way steel does, which affects arc stability and how much heat actually lands in the joint.

A Narrow Window Between Solid and Liquid

Because heat escapes so quickly, aluminum transitions from “not hot enough” to “a hole in your workpiece” in a hurry. The margin for error is thin, and the base metal offers few visual warnings. Steel glows red before it melts and gives you a heads-up; aluminum simply goes from silver to a shimmering puddle with almost no color change, so you learn to read the puddle surface and its motion instead. Add a low melting point relative to steel and you have a material that punishes a slow hand.

Expansion, Distortion, and Built-In Stress

Aluminum also has a high coefficient of thermal expansion. It grows noticeably as it heats and shrinks as it cools, pulling and warping more readily than steel. Weld it restrained and that movement turns into stress, which is a direct recipe for cracking.

Quick Note: These same properties are why so many welders treat aluminum like “shiny steel” and fail. Steel habits — slower travel, tighter fit-up, no oxide removal — all work against aluminum’s grain.

Why This Matters Before You Strike an Arc

Every property here has a matching technique:

Oxide layer → clean immediately before welding, and use AC or reverse polarity to break it up.
High conductivity → crank the heat and expect to move faster.
Small melting window → practice, because there’s little room to react.
Thermal expansion → tack well, sequence your passes, and manage restraint.

None of this clicks from reading alone. It clicks at the bench, which is why deliberate welding practice on scrap aluminum pays off faster than any settings chart. For a deeper dive into arc behavior, the Aluminum Extruders Council’s guide to aluminum fusion welding is a solid reference, and AWS maintains the D1.2 Structural Welding Code for Aluminum that governs work on critical parts.

Reading the Bead: What a Good Aluminum Weld Looks Like

The moment you lift your hood, you have a wealth of information in front of you — if you know what to look for. An aluminum bead talks. It tells you whether your cleaning action worked, whether your gas coverage held, and whether your travel speed stayed steady. Plenty of ugly beads are actually sound, and plenty of pretty ones hide defects. Here’s how to separate the two.

Bead Shape and Convexity

A good aluminum bead sits slightly crowned — gently convex, not lumpy. Think of a shallow loaf, not a rope. On a fillet weld the face should be flat to mildly convex, with no sharp shelf where it meets the base metal. Excessively concave fillets are the danger. They fail through the effective throat because there simply isn’t enough metal at the root, and they’re unacceptable under AWS weld profile criteria. A bead that sags or bulges unpredictably signals inconsistent travel speed or a pool you weren’t controlling.

The Ripple Question: Stack of Dimes or Smooth?

Everyone has heard the “stack of dimes” praise, but uniformity matters more than style. Good ripples are evenly spaced and consistent in height along the whole length — a rhythm you can read from one end to the other. The pattern itself is a cosmetic cue, not a soundness test. A smooth, near-ripple-free bead from a steady autogenous pass can be perfectly sound, while a gorgeous stack of dimes can still hide internal porosity. Judge consistency first, aesthetics second.

The Cleaned Zone: Your Best Visual Clue

Look just outside the bead. In AC TIG welding, the electrode-positive phase produces the surface cleaning action that breaks up the aluminum oxide layer. A bright, frosted “cleaned” band alongside the weld means that action worked and your shielding gas protected the pool. Master that balance by understanding TIG welding polarity. A dull, gray, oxidized bead with no cleaned zone tells you shielding failed, gas flow was wrong, or you traveled too fast.

Quick Note: A frosted halo that’s even on both sides says your torch angle and gas flow are right. A halo that’s lopsided, or missing entirely, is an early warning you can fix before the next pass.

Toes, Crater, and the Sooty Telltales

Run your eye down the toes — where the bead meets the base metal. They should blend smoothly, no undercut, no overlap bead hanging over the edge. A crater at the end that’s a shallow dimple is fine; a deep, cracked pit is not. Then look for the trouble signs: heavy black sooty smut (usually a shielding gas or contamination problem), spatter scattered around the bead, undercut, or a concave cap. Mechanical pores and elongated porosity show up most often where shielding broke down or travel speed ran too high.

The Tactile Check

Once the metal is cool, run a fingertip along the toes. A good weld feels like one continuous surface; your nail should catch on nothing. A wire brush that snags means undercut or an overlap defect. Tap the cooled bead with a light hammer — a clean ring beats a dull thud, which can hint at a void or incomplete fusion underneath. This is the habit that separates people who just run beads from those who build real welding practice.

Cosmetic Cues vs. Real Soundness

Here’s the honest truth: your eyes can’t see inside the metal. A bead can look textbook and still be weak. Treat appearance as a screening tool, not a verdict, then confirm with the inspection methods that actually peer below the surface. If you weld MIG, compare your results against our guide to good and bad MIG welds, and remember that consumables like a worn contact tip can throw off the bead before technique ever becomes the issue.

Cleanliness and Preparation: The Foundation of Every Good Aluminum Weld

Ask a veteran aluminum welder where good welds come from and they’ll usually point at the bench, not the torch. Aluminum welding is roughly 80% preparation and 20% arc time. Skip the prep and no machine setting on earth will save you. The metal answers back with porosity, soot, and cracks.

Stage One: Degrease Before You Touch the Surface

Start by removing oils, greases, cutting fluid, marker ink, and adhesive residue with acetone or isopropyl alcohol. These hydrocarbons are major hydrogen sources, and hydrogen is the number-one cause of porosity in aluminum. Use clean, lint-free cloths and wipe until they come away white. Don’t reuse a solvent-soaked rag that’s already been across a steel plate, since that just smears contamination around.

Critically, degrease before you brush. Scrub the surface first and you push oil and grease down into freshly opened metal, where no amount of later wiping can pull it back out.

Stage Two: Mechanically Break the Oxide

Aluminum instantly grows a tough oxide layer that melts around 3,700°F (2,050°C), far higher than the metal beneath it, which melts near 1,220°F (660°C). That skin has to come off mechanically. Use a dedicated stainless steel wire brush or a carbide burr, and run the tool in one direction only. Back-and-forth strokes can work broken oxide and abraded debris right back into the surface.

Many welders assume the AC arc’s cleaning action handles this for them. It helps, but it never replaces prep. To understand what that arc is really doing, read our guide to TIG welding polarity: DCEN, DCEP, and AC.

The Clock Is Ticking: Weld Within One to Two Hours

Aluminum oxide reforms almost immediately after brushing, and a hydrated oxide layer keeps pulling moisture from the air. Weld within one to two hours of cleaning for reliable results. If a part sits overnight in a humid shop, clean it again. There’s no shortcut.

That moisture matters because of metallurgy. Molten aluminum dissolves far more hydrogen than the solid metal can hold, so as the weld freezes, the excess hydrogen is rejected at the solidification front and trapped as gas bubbles. Those are the pores you find later.

Your shielding gas supply matters too. Aluminum welding calls for a minimum purity of 99.995% and a dew point of −60°C or lower, as outlined in the ESAB aluminum technical guide. A wet or contaminated cylinder pushes hydrogen straight into the arc.

Quick Note: Hydrogen is the enemy here. Moisture, hydrocarbons, and hydrated aluminum oxide all feed it into the weld, which is why clean, dry metal and dry gas matter so much.

Back-Gouging and Back-Chipping for Full Penetration

On full-penetration joints, cleaning the top is only half the job. The root and back side hide contamination that will rise up through the puddle. Back-gouge with a carbide burr or grind back to sound metal, then re-clean before the next pass. Back-chipping removes the unsound root and any oxide formed during the first pass, giving the sealing pass clean metal to fuse into.

Keep Aluminum Tooling Away From Steel

Steel brushes, grinding wheels, and files carry rust, mill scale, and oil, all contaminants for aluminum. Set aside aluminum-only tools and store them where they won’t get mixed in with steel gear. A well-organized shop makes this easy; our guide to planning and setting up a safe welding facility covers the layout side.

If your beads still wander after all this, the fix is usually seat time, not settings. Our guide to welding practice walks through drills that build real consistency.

Contamination Sources to Hunt Down

Beyond oil and grease, watch for:

Paint, primer, and coatings that trap solvents and moisture
Anodized or heavily oxidized surfaces
Moisture from condensation, rain, or humid storage
Shop dust, grinding grit, and airborne steel particles
Marker ink, tape residue, and fingerprints

Store aluminum dry and covered, off bare concrete, and handle it with clean gloves. Every one of these contaminants feeds the same hydrogen-porosity problem that shows up later in this article, so treat prep as the job itself, not the warm-up.

Filler Alloy and Base Metal Selection: Getting the Chemistry Right

Pick the wrong filler and even a flawless-looking aluminum bead can crack as it cools or crumble years later in service. Base metal chemistry and filler chemistry have to work together, and the acceptable window is narrower than most welders expect. The difference between a joint that lasts and one that fails often comes down to the filler you reach for. Everything starts with the alloy series stamped on the material you’re joining.

The Base Alloy Series Sets the Rules

Non-heat-treatable alloys — the 1xxx, 3xxx, 4xxx, and 5xxx families — are generally the easiest to weld. They gain strength through work hardening rather than heat treatment, so welding heat doesn’t scramble their metallurgy. Heat-treatable alloys (2xxx, 6xxx, and 7xxx) are a different story. They harden through alloying elements that precipitate out, and the same chemistry that makes them strong also leaves them sensitive to hot-short cracking.

The mechanism shows up in the hot-cracking curves. These curves plot crack susceptibility against filler dilution — how much of the weld pool is melted base metal versus added filler. For crack-sensitive alloys a dangerous zone exists, and the fix is to dilute your way out of it. The rule of thumb: the weld metal should consist of at least 70% filler alloy so the final chemistry lands outside the crack-sensitive range. That’s the coefficient-of-dilution concept in practice. The more base metal you melt into the pool, the more you drag the chemistry toward trouble.

4043 vs 5356: The Two Workhorses

Two fillers cover the vast majority of aluminum work: 4043 and 5356. The choice comes down to service conditions more than personal preference, as ESAB’s six filler-selection factors make clear.

Property	4043 (Al-5Si)	5356 (Al-5Mg)
Melting / fluidity	Lower melting point, very fluid	Higher melting point, stiffer
Strength & ductility	Moderate	Higher strength, better ductility
Corrosion	Good	Excellent, marine-grade
Anodized color	Dark, near-black	Lighter, closer to base
Best use	Most 6xxx, crack-prone joints	Where strength matters

4043 (Al-5Si) carries silicon, which lowers the melting point and boosts fluidity. The result is smoother beads, better wetting, and natural resistance to cracking — which is why it’s the preferred filler for many 6xxx applications, including 6061. Its downside is a dark anodized finish.

5356 (Al-5Mg) is magnesium-bearing and delivers higher strength plus noticeably better ductility. It also resists corrosion strongly in marine and saltwater settings and anodizes to a lighter tone that often matches the base metal better. The catch is temperature: 5356 is not recommended for sustained service above 150 °F (65 °C), because Al2Mg can precipitate at grain boundaries and open the door to stress-corrosion cracking.

Why 6061 Needs a Dissimilar Filler

Here’s the counterintuitive part. 6061-T6 arrives with roughly 45,000 psi tensile strength, but the as-welded joint typically drops to about 27,000 psi — the heat treatment that provided the strength is partly undone by welding heat. You don’t recover that by matching chemistry. Welding 6061 with 6061 filler actually pushes the pool into a crack-sensitive range. The fix is a dissimilar filler, usually 4043 or 5356, that shifts weld-metal chemistry away from the hot-cracking zone while treating the strength loss in the heat-affected zone as normal and predictable. In other words, the filler is there to steer chemistry, not to match the base metal. Need the properties back? Post-weld heat treatment, or a smarter joint design, is the real answer.

The Alloys That Fight Back: 2xxx and 7xxx

The 2xxx (copper-bearing) and 7xxx (zinc-bearing) families are aerospace grades — very strong and very difficult to weld. Their chemistry sits deep inside the crack-sensitive range, and several grades are considered effectively non-weldable by fusion. If you must join them, you’ll need specialized fillers, preheat, and a procedure qualified to a recognized standard such as AWS D1.2. For most shops, mechanical fastening, riveting, or adhesive bonding is the smarter route.

Quick Note: The 70% filler rule isn’t a suggestion — it’s the buffer that keeps crack-sensitive weld metal out of the danger zone. If your joint melts in a lot of base metal (thin parts, deep penetration), switch to a filler with better crack resistance.

None of this matters if the arc can’t deliver a stable pool. Process and polarity choices shape how much base metal dilutes into the weld, so it’s worth understanding the related perspective on DC welding aluminium when a job refuses to behave. If you’re still building the fundamentals of bead control, our guide to real welding practice covers the hands-on habits that make filler selection pay off.

Process and Settings: AC TIG, MIG, Polarity, Gas, and Technique

Aluminum does not forgive sloppy machine settings. Get the process, polarity, gas, and travel speed right and the metal practically welds itself. Get them wrong and you fight a sputtering arc, a dirty puddle, and a bead that cracks an hour later. Here is how the two main processes are dialed in for sound aluminum welds.

AC TIG: The Standard for Aluminum

TIG is the go-to process for thin aluminum and any joint where appearance matters, and it works on AC for one reason: aluminum wants two contradictory things from the arc at once.

Electrode negative (EN): the work is positive, so heat pours into the aluminum. This half of the cycle drives penetration.
Electrode positive (EP): the work is negative and the arc bombards the surface, breaking up the tough aluminum-oxide layer. This is the cleaning half.

AC alternates between the two, giving you penetration and cleaning in a single pass. Your balance control decides the split. A common starting point is about 70% EN and 30% EP: enough cleaning to keep the puddle bright, enough penetration to fuse the joint. If the oxide fights you, push toward more EP; if you are burning through thin stock, bias back toward EN. Frequency controls how tightly the arc focuses: higher frequencies narrow the arc and concentrate heat, while lower frequencies widen it and soften the cleaning. Most machines give you a range of roughly 60 to 120 Hz, and starting near 100 Hz is a fine middle ground for general work.

For amperage, a rough starting rule is about 1 amp per 0.001 inch of material thickness. So 1/8 inch aluminum (0.125 in) starts around 125 amps. Seasoned welders bend that rule based on joint type, position, and preheat, so treat it as a baseline rather than gospel. The full breakdown of how AC, DCEN, and DCEP behave lives in our TIG polarity guide, and if you are still chasing consistency, structured welding practice does more than any setting tweak. These fundamentals echo the AWS aluminum welding guidance that shop manuals are built on.

Post-flow matters more than most people expect. Keep shielding gas flowing after the arc stops to protect the cooling tungsten and puddle; up to about 10 seconds is normal on thicker work. If you pull your tungsten and it is discolored, your post-flow is too short.

Quick Note: On thin aluminum, most of the heat you add becomes distortion. A tight arc, fast travel, and a slight preheat on thicker parts all cut heat input where it counts.

MIG Settings for Aluminum

MIG (GMAW) is faster and more forgiving on thicker material, but soft aluminum wire brings its own headaches. Pure argon is the most widely used shielding gas for aluminum in both GMAW and GTAW because of its better arc starting, cleaning action, and stability. Helium-argon mixtures, commonly between 30 and 70% of each, raise arc power, deepen penetration, reduce porosity, and let you travel faster. Heavier helium-rich mixes need higher gas flow rates than argon because helium is much less dense and escapes the puddle area faster. As a baseline, run argon around 15 to 25 cfh and push helium mixes toward the upper end of that range.

Technique on aluminum is push, not drag. Pushing the gun keeps the shielding gas ahead of the puddle and gives you a clearer view of the joint. Hold a short contact-tip-to-work distance, roughly 3/8 to 1/2 inch, and keep the gun angle nearly perpendicular, tilted only about 10 to 15 degrees in the direction of travel. Too much stickout stresses the arc; too little clogs the nozzle. Our contact tip welding guide covers selection, wear, and the feeding faults that show up at the tip.

Soft wire feeds like cooked spaghetti, so short, straight gun leads help. For thinner wire or long reach, a spool gun or push-pull feeder keeps the wire from bird-nesting. Set drive-roll tension just tight enough to feed without crushing the wire; our wire welder beginner guide walks through feeder setup.

On thick sections, preheat to about 150–200°F (roughly 65–95°C) to improve fusion and let you weld at lower amps. Then travel fast. High travel speed limits heat input, which is your main weapon against distortion and hot cracking. Some shops use DC electrode negative for specific aluminum jobs; DC welding aluminium explains when that helps.

Defects That Ruin Aluminum Welds: Porosity, Cracking, and Lack of Fusion

Aluminum welds fail in predictable ways. Lack of fusion and porosity are considered the two most common discontinuities in aluminum welds, so they deserve top billing. Learn why each one forms and you’re most of the way to preventing it.

Hydrogen Porosity and Shrinkage Porosity

Hydrogen is the number one cause of aluminum porosity. Molten aluminum dissolves far more hydrogen than the solid metal, and as the weld freezes, that hydrogen has nowhere to go. It forms bubbles that get trapped before the metal solidifies. The culprits are almost always contamination: moisture, oil, grease, hydrated surface oxide, or humid shop air feeding hydrogen into the arc. Excessive hydrogen porosity reduces strength, ductility, and fatigue resistance by shrinking the load-bearing cross-section and creating stress concentrations.

A second, subtler type is inter-dendritic shrinkage porosity. Here the voids aren’t gas — they’re gaps left where the last liquid to freeze couldn’t feed the shrinking dendrite network. It clusters near the centerline and in wide, slow-freezing beads, especially with high heat input and no backing.

Fixes: clean the joint to bright metal and weld within hours, use the right filler, control travel speed, and cap heat input.

Quick Note: Aluminum stored in a humid shop absorbs moisture into its oxide layer. In muggy summer months, a part that welded clean in February can turn porous in July. Wipe with a dedicated stainless brush and degreaser right before welding, and don’t leave prepped parts overnight. A dry, ventilated welding facility makes a real difference here.

Solidification Cracking and Crater Cracks

Solidification cracking occurs in the last stage of freezing when tensile residual stresses exceed the strength of the partially solidified weld metal. It runs hot, straight down the centerline. The 6xxx alloys are the classic offenders: 6061 is notorious for high solidification-cracking susceptibility because it forms an insufficient volume of terminal eutectic liquid to backfill the grain boundaries as they freeze.

Grain structure decides how bad it gets. Fine equiaxed grains are less prone to solidification cracking than coarse columnar grains, because their boundaries are more tortuous and better wetted by liquid. Titanium and zirconium in fillers like 5356 act as grain refiners, promoting exactly that structure (aluminum fusion welding practice). Choosing 4043 or 5356 over a matching 6061 rod, plus sensible joint design, helps a lot (filler selection guide).

Crater cracks are a close cousin. Crater cracks often form because the crater has a smaller cross-section and does not receive enough filler to shift chemistry away from the crack-sensitive range. Fix: taper the current and add filler as you finish, or use a run-off tab.

Lack of Fusion and Incomplete Penetration

Lack of fusion happens when the arc doesn’t melt the sidewalls, leaving a cold lap the next layer never bonds to. Incomplete penetration means the root didn’t melt through. Both usually trace to too little current, the wrong polarity, fast travel speed, or poor fit-up. Raising amperage, slowing down, beveling the joint, and confirming your polarity fix most cases.

Oxide and Tungsten Inclusions

Aluminum oxide melts well above the aluminum beneath it. If the arc doesn’t break it up, oxide films get folded into the puddle as inclusions. On AC TIG, insufficient cleaning action — too much electrode-negative balance — leaves that skin behind. Tungsten inclusions come from dipping the electrode or overcurrent, showing up on radiographs as bright, sharp specks. Correct AC balance, a properly ground electrode, and steady hands solve both.

Distortion

Aluminum conducts heat roughly five times faster than steel and expands more, so it warps easily, and excess heat invites stress-concentrating undercut. Minimize heat input, use backstep welding and balanced sequences, tack strategically, and clamp where you can.

None of these flaws are mysterious once you know the metallurgy. For a related family of problems, see our breakdown of the overlap welding defect, and check AWS D10.7M for acceptance criteria.

Inspection and Testing: How to Verify a Good Aluminum Weld

A bead that passes the eye test can still fail a bend. Soundness is proven, not assumed, and the proof runs from a flashlight and fillet gauge to radiography and guided-bend jigs.

Visual Inspection Comes First

AWS D1.2 treats visual examination as the primary acceptance gate, and it is where most defects get caught cheaply. Check bead profile for uniform ripple, width, and crown height. Watch undercut depth against thickness-based limits, and confirm the crater is filled, not left as a crack seed. Any crack visible on the surface is rejectable outright. Also confirm leg size, reinforcement height, and that the weld toes blend smoothly into the base metal without a sharp notch.

Quick Note: Good lighting and magnification reveal more than the naked eye. A 10x loupe costs little and catches toe cracks, porosity, and incomplete fusion that a casual glance misses.

Radiographic, Penetrant, and Ultrasonic Testing

AWS D1.2 outlines radiographic and dye-penetrant testing as standard volumetric and surface methods. Dye penetrant (PT) will expose surface-breaking porosity and cracks, while radiography (RT) images internal porosity, inclusions, and lack of fusion. Ultrasonic testing is permitted for aluminum, but the procedure and acceptance criteria must be written into the contract documents; UT is not a default substitute for RT. In practice, PT is the fast surface check on finished parts, RT is reserved for critical joints where internal soundness matters, and UT shines on thicker sections and T-joints that radiography struggles to image.

Mechanical Testing: Bend, Macroetch, Tensile

Bend, macroetch, and tensile tests confirm what the part can actually do. Aluminum bend testing differs from steel because the material is softer and more ductile, so it tears and stretches differently than carbon steel. That is why the preferred method is a wraparound guided bend jig per AWS D1.2, which supports the specimen through the wrap and reveals porosity, lack of fusion, and coarse grain. Tensile testing pulls a coupon until it breaks, confirming the joint reaches the minimum specified strength rather than the base metal’s full value. Macroetching cuts, polishes, and etches a cross-section to expose penetration depth, root fusion, and internal porosity in a single look.

Qualified Procedures and Welders

None of this testing means much without a qualified welding procedure specification. Every joint detail and procedure under AWS D1.2 must be individually qualified and captured in a WPS backed by a procedure qualification record (PQR), with welders qualified to that procedure. Documentation is not bureaucracy here; it is the only proof that a weld was made to a qualified procedure.

Quick Shop-Level Checks

Most of what gets rejected could have been caught in the first five minutes after the arc stopped. Try these:

Tap and flex a sample; a sharp ring or clean bend beats a dull thud.
Run a wire brush over suspect areas and re-inspect for cracks.
Gauge crown height and undercut instead of trusting the eye.
Cut a scrap coupon and macroetch the cross-section to see penetration and porosity.
Keep the qualified WPS and welder certificates on file so inspection paperwork follows the part.

Filler choice shows up here too: X-ray CT studies found about 18.3% porosity by area in AA6061 welds made with ER4043 versus roughly 8.4% with ER5356, so a switch in filler can move you from reject to pass.

Testing catches problems, but consistency comes from practice. Sharpen your technique with our welding practice guide before you trust a bead to a bend jig.

Common Mistakes and Field-Tested Fixes for Better Aluminum Welds

Everything up to this point is theory. This is the part you tape to the shop wall. Most bad aluminum welds trace back to a short list of repeatable mistakes, and nearly every one has a simple, proven fix. The good news is that once you recognize the pattern, correcting it becomes second nature.

Stop Treating Aluminum Like Steel

The single biggest error is running aluminum through a steel mindset. Steel is forgiving; aluminum is not. It oxidizes within seconds, pulls heat away far faster, and gives no color warning before it collapses into a puddle. That means sharing equipment, brushing habits, and technique designed for carbon steel is asking for trouble.

Welding dirty metal. Aluminum oxide melts at roughly 3,700°F, over three times the melting point of the base metal. Wipe with a dedicated solvent, then remove the oxide mechanically right before welding.
Using a steel-dedicated brush. A stainless steel brush used in one direction only prevents embedding steel contaminants into the soft aluminum surface. Never use a brush that has touched steel, and never scrub back and forth.
Wet or contaminated filler wire. Store filler in a sealed tube. Moisture on the wire is a direct hydrogen source, and hydrogen is what creates porosity.
Too-long arc and dragging the MIG gun. Push, don’t drag, and keep the arc tight. A long arc loses shielding, overheats the wire, and invites spatter and lack of fusion.
Wrong filler or over-dilution. Choosing filler by habit instead of by base alloy weakens the joint. Match chemistry deliberately, and avoid burning back too much base metal into the puddle.

Quick Note: Many fabricators notice welding problems spike with the seasons. Airborne moisture is a prime hydrogen source, so humidity swings often show up as sudden porosity in otherwise identical work. Drier filler, drier airflow, better results. Setting up a clean, controlled work area goes a long way — see our guide to planning and setting up a safe shop.

Fix Feedability Problems

Soft aluminum wire kinks and shaves far more easily than steel. Fit U-groove drive rolls, which cradle the wire instead of biting into it and reducing shavings. Keep liner runs short and gentle, and dedicate a torch and liner to aluminum whenever the budget allows. If you switch back and forth, at minimum use dedicated wire guides so steel residue never contaminates the feed path. A worn or wrong-sized tip will wreck an otherwise good setup, so it pays to understand contact tip selection and maintenance.

Keep Distortion From Running Away

Aluminum expands and contracts hard, and thin sections warp fast. Because it cools so quickly, the shrinking forces arrive almost before you can react. Control it before you strike an arc:

Use balanced welding sequences and distribute welds symmetrically around the part.
Weld with the correct size — oversize beads add heat and shrink forces.
Use intermittent (stitch) welds where the joint allows, and make fewer beads overall.
Fixture or clamp the workpiece to hold geometry while it cools.

Poor fit-up deserves its own mention. Gaps force you to fill instead of fuse, which overheats the part and multiplies distortion. Tack accurately, then check square before committing.

Insufficient Post-Flow and Skill Gaps

Cutting shielding gas the instant you release the trigger leaves a molten, unprotected puddle that sucks in air. Hold post-flow long enough for the weld to freeze. And honestly assess your technique: if beads wander, run structured practice drills until travel speed, torch angle, and rhythm become automatic. Understanding your machine’s polarity settings for TIG also clears up a lot of confusing defects.

Build the habit loop: clean, dedicated tooling, tight arc, conservative heat, controlled cooling. Log what works for each joint and alloy, because the variables shift from job to job. Do that consistently and good aluminum welds stop being luck.

Conclusion: Building Consistently Good Aluminum Welds

By now the pattern should be obvious: good aluminum welds are not accidents. They are the predictable output of a routine you run the same way every single time. The metal changes, the joint changes, but the process does not.

That process starts before the arc ever strikes. Know your base alloy and your filler alloy. Scrub the oxide off and weld before it returns. Dial in polarity, gas flow, and amperage. Then treat heat input and travel speed as one decision, not two, because aluminum takes a great deal of heat to melt and too much heat per inch will pull a thin panel into a banana before you ever lift your hood.

The end result depends on four things working together: proper cleaning, correct filler selection, a stable shielding gas shield, and appropriate welding parameters. Miss any one and the other three cannot save you. That holds whether you run AC TIG, spray-transfer MIG, or a DC setup, which is why a solid grasp of TIG welding polarity pays off in every process you touch.

Notice how little of this depends on your hand. The puddle follows the prep, the chemistry, and the settings; once those are right, your job is simply to move steadily, keep the arc short, and stop before the part overheats.

A Short, Repeatable Action Plan

Identify the alloys. Match filler to base metal before you strike an arc.
Clean, then clean again. Degrease, stainless-brush or etch, then weld fast.
Set up deliberately. Polarity, gas, amps, and travel speed chosen on purpose.
Watch the bead and the crater. A cracked crater is a warning shot.
Verify, do not hope. Visual first, then penetrant, bend, or X-ray where it counts.

Reading the bead is your fastest feedback loop. A sound aluminum weld sits smooth and evenly rippled, wets both toes cleanly, and shows no soot or pinholes. When something looks off, fix the cause rather than the symptom. A sooty MIG bead usually points to shielding trouble, not wire trouble, and knowing good and bad MIG welds carries straight over to aluminum. Keep the contact tip fresh and the liner clean, because wire-feed stutter can ruin an otherwise perfect bead.

On structural work, all of this stops being a personal habit and becomes a requirement. Under AWS D1.2, procedures and personnel must be qualified before production welding begins. A written WPS is not paperwork for its own sake; it is how you make one good weld reproducible a thousand times.

Quick Note: Fillers are matched to base alloys using established selection guides, and the chemistry determines crack resistance as much as your hand does. Get the alloy right before you blame your technique.

None of this is raw talent. Consistent aluminum welding is a learnable skill, and like every weld skill it is built through deliberate, repetitive practice — the kind that builds muscle memory for steady travel speed and a controlled dip. Resources like this welding practice guide can structure that work so each session builds on the last.

So treat quality as a process, not a single bead. Run your routine, trust your inspection, and let each weld teach the next one. Consistency is the prize, and it is yours to build, one weld at a time.

Frequently Asked Questions

What does a good aluminum weld look like?

A good aluminum weld has a smooth, uniform bead with a slightly convex profile, consistent ripples, tight tie-in at the toes, and a bright, frosted ‘cleaned’ band where the arc broke up the oxide. It should show no undercut, no soot, minimal spatter, and a properly filled crater. Looks alone are not enough, though — soundness also depends on fusion and freedom from porosity and cracks.

Why is aluminum harder to weld than steel?

Aluminum has a tough oxide layer that melts around 2000 °C while the metal melts near 660 °C, so the oxide stays solid in the pool. It also conducts heat about three times faster than steel, expands more when heated, and has a narrow welding window. Together these traits promote porosity, cracking, lack of fusion, and distortion.

Should I use 4043 or 5356 filler for aluminum?

Use 4043 (Al-5Si) for many 6xxx base alloys because it cracks less, flows better, and performs at higher service temperatures. Use 5356 (Al-5Mg) when you need more strength and ductility, especially on 5xxx alloys, but avoid it for sustained service above 150 °F (65 °C) because of stress-corrosion cracking risk.

How do I prevent porosity in aluminum welds?

Porosity is almost always hydrogen from moisture, oil, grease, or hydrated oxide. Degrease with acetone, brush off the oxide with a dedicated stainless steel brush, weld within one to two hours, use clean dry filler wire, and ensure high-purity shielding gas (99.995% or better) with no leaks in the system.

Can you MIG weld aluminum without a spool gun?

Yes, but feeding soft aluminum wire through a long standard torch is difficult. Spool guns, push-pull feeders, short liner runs, and U-groove drive rolls all help. Dedicated liners and guides for aluminum prevent contamination and bird-nesting.

Why does my aluminum weld crack right down the center?

Centerline cracking is solidification (hot) cracking, driven by the chemistry of the base and filler combination plus restraint stresses. Crack-sensitive alloys like 6061 need enough dissimilar filler — generally at least 70% filler in the weld metal — to move the chemistry out of the crack-sensitive range.

What polarity is best for TIG welding aluminum?

AC is the standard for aluminum TIG because it alternates between electrode-negative for penetration and electrode-positive for oxide cleaning. A common starting balance is about 70% EN and 30% EP, adjusted for how much cleaning the material needs.

How do I know if my aluminum weld is actually sound?

Inspect visually against AWS D1.2 criteria for profile, undercut, and crater condition, then use radiographic or dye-penetrant testing where specified. For structural work, the weld must follow a qualified WPS and be made by a qualified welder. Mechanical tests like guided bend and macroetch confirm fusion and soundness.