Melt-Through Weld Defect: Causes, Prevention, and Repair (Complete Guide)

What Is Melt-Through? The Weld Defect That Hides in Plain Sight

You are running a beautiful bead on a thin stainless tube. The arc sounds perfect, the puddle glows, and then — a glowing droplet of molten metal drops out the bottom and leaves a ragged hole. That sinking feeling has a name. Melt-through is one of the most common and most preventable weld defects in thin-section work, and it often shows up after the damage is done.

In plain terms, melt-through occurs when heat input is high enough to fully penetrate the joint and push molten weld metal through to the root side. The result is a sagging bulge, a drip, or an open void on the underside of the weld — even when the cap looks acceptable from above. That gap between a clean-looking face and a damaged root is why this defect hides in plain sight.

The confusion starts with vocabulary. Welders, inspectors, and code documents do not always use these terms alike, so it is worth untangling three that sound almost identical.

Melt-Through vs. Burn-Through vs. Excess Root Penetration

AWS A3.0M/A3.0:2025, Standard Welding Terms and Definitions (14th edition, ANSI-approved July 30, 2024) groups “Melt-Through and Root Surface Profile” together in Appendix B.27 — a hint that the two ideas are closely related. On the shop floor, the shorthand is looser:

• Melt-through usually describes excessive molten metal passing through the joint to the root side, leaving a bulge, sag, or drip.
• Burn-through is the more severe cousin: molten weld metal drops completely through the joint and forms an underfilled bead, with a concave top surface and a convex underside, as ScienceDirect’s welding-defect overview describes.
• Excess root penetration is controlled, uniform penetration that goes beyond what the drawing or code allows — no hole, just more root reinforcement than specified.

In other words, melt-through and burn-through sit on a sliding scale of “too much heat, too little material,” and the boundary between them is usually about severity, not a hard line in the sand.

It is also easy to misread melt-through as a cause rather than a result. Industry references list melt-through among defect causes alongside lack of shielding gas, oily parts, a mislocated torch, and large root openings — all of which point back to one theme: too much heat concentrating in too little metal.

Why Melt-Through Matters More Than It Looks

A small drip under a thin stainless tube may look cosmetic, but the consequences rarely are:

• Fatigue life — a sharp-edged hole or undercut acts as a stress riser and can crack under cyclic loading.
• Leak paths — any through-thickness void in a pressure or fluid line is a future leak, not a maybe.
• Corrosion — a rough, oxidized root and crevices trap moisture and accelerate attack, especially in stainless.
• Rework cost — cutting out and re-welding thin material often costs far more than the original joint.
• Code rejection — if the joint fails visual or radiographic acceptance criteria, the whole assembly can be rejected.

If you are still building the fundamentals, our guides on overlap welding defects and on telling good and bad MIG welds apart are useful companions — melt-through is often the final symptom of settings and travel-speed issues that started long before the arc struck.

What This Guide Covers

This article walks through the full picture: how melt-through differs from neighboring defects, the physics of too much heat, how it behaves across TIG, MIG, stick, flux-core, and orbital welding, how to spot it with visual and NDT inspection, a prevention playbook covering settings, technique, fixturing, and purge control, proven repair methods, the challenges of aluminum, stainless, and titanium, and what the codes say about acceptance. By the end, you will know how to avoid a hole — and exactly what to do when one appears.

Melt-Through vs Burn-Through vs Excess Penetration vs Lack of Penetration

Welders use these four terms like synonyms. They are not. Mix them up on a WPS or a repair ticket and you will fix the wrong problem — or make a bad joint worse.

The picture is simple: two plates, a V-groove, one weld. The root is the bottom of the joint, and how metal behaves there decides which word applies.

The Four Terms in Plain Language

• Melt-through: heat is high enough that the root melts open and molten metal drops out the far side, leaving a ragged hole or a hanging droplet.
• Burn-through: the next step up — a full hole blown clean through the part. Melt-through warns you; burn-through is the failure.
• Excess penetration: the root fuses properly, but reinforcement hangs too far past the root surface. Too much of a good thing.
• Lack of penetration (incomplete joint penetration): the opposite failure — weld metal never reaches the root. AWS B1.10M/B1.10:2016 defines it at clause 4.6 as a joint root condition where weld metal does not extend through the joint thickness.

Side-by-Side Comparison

Reading the Root Surface Profile

Inspectors describe the root by its surface profile, and the words carry weight. A flush root sits level with the surface. A convex root bulges outward; keep bulging and you get excess penetration. A concave root dips inward — a sign of too much heat or insufficient fill, and often the first stage of melt-through. Hang a droplet of metal off the root and you have “icicles” from drop-through. On stainless you may also see sugaring, the rough, oxidized scale that forms when purge gas fails.

Why Vocabulary Matters for WPS/PQR Work and Repairs

AWS B1.10 goes further than definitions: it separates a discontinuity from a defect. A discontinuity is any interruption in the normal structure; it becomes a defect only when a code or standard says it is unacceptable. Excess penetration below the limit is a discontinuity you document. Above the limit, it is a defect you repair. That distinction drives real decisions — a repair disposition has to name the condition correctly, because the fix for lack of penetration (more heat, tighter fit-up) is the exact opposite of the fix for melt-through (less heat, faster travel). It also matters in WPS/PQR development, where each essential variable must target the right phenomenon. Get the term wrong and you qualify the wrong procedure.

AWS A3.0, the standard welding terms and definitions, is the shared dictionary here: when a drawing says “excess penetration” or “incomplete joint penetration,” that is the language being referenced.

Fit-up is the hidden variable in all four conditions, especially on tubing where an oversized gap funnels heat straight to the root. Our guide to weld tube fittings and fit-up covers how to control that gap before the arc ever starts. For the practical side of reading a bead and telling good from bad, see good and bad MIG welds. And because consistency comes from repetition under guidance, structured welding practice is what makes these distinctions second nature. The science behind all of it — heat input, arc energy, and thermal gradients — is explored in welding defect research.

The Physics of Too Much Heat: Why Melt-Through Happens

Melt-through is not bad luck — it is arithmetic. Every arc pumps a measurable amount of thermal energy into the joint, and when that energy outruns what the surrounding metal can absorb and conduct away, the pool burns a hole clean through. Once you can see the equation behind “running hot,” you stop guessing and start controlling it.

Heat Input: The Number Behind the Burn

The Polysoude Orbital Welding Handbook reduces it to one line:

HI = (60 × U × I) / S

• HI — heat input, in J/mm or J/in
• U — arc voltage (volts)
• I — welding current (amps)
• S — travel speed (mm/min or in/min)

Look at the structure. Volts and amps sit on top; travel speed sits underneath. Raise either electrical value and heat input climbs. Slow the travel speed and it climbs again, because the same arc pours energy into the same millimetre of metal for longer. The 60 is just a unit conversion constant; the physics lives in the ratio.

The Variables That Push Heat Input Over the Edge

Melt-through is rarely one setting gone wrong — it is several small excesses stacking up:

• Amperage too high. Current does most of the work, and too much of it is the classic trigger. Amperage, travel speed, and dirty base metal are named as the primary burn-through causes by industry trainers like UNIMIG.
• Voltage too high. Voltage tracks arc length; a long arc spreads heat, widens the pool, and overheats thin material.
• Travel speed too slow. As the formula shows, dwelling is the quiet killer — especially at starts, stops, and joint edges where we instinctively hesitate.
• Dwell at the edges. That pause to “tie in” a corner dumps extra joules into an already-thin wall.
• Wrong polarity. Electrode-positive (DCEP) concentrates more heat in the workpiece. Great for cleaning, brutal on thin gauge — see our breakdown of TIG welding polarity: DCEN, DCEP, and AC.
• Poor fit-up. A root gap wider than specified means more open metal to melt with no support behind it.
• Insufficient backing. No chill bar, purge, or backing gas and the root has nothing to hold the pool.

Arc length ties into consumable condition too: a worn nozzle or a contact tip with a damaged bore gives an unstable, wandering arc and inconsistent penetration.

Why Aluminum Fights Back

Aluminum is the poster child for melt-through. It melts at roughly 660°C, yet its tenacious surface oxide melts near 2000°C — three times hotter than the metal underneath. The oxide insulates the molten puddle and starves the arc of a clean, stable path, so welders crank the amperage to cut through the skin. The moment it breaks, all that heat hits metal that was already ready to collapse. Add aluminum’s thermal conductivity — about six times that of steel — plus its low melting point, and you have a material MillerWelds identifies as highly susceptible to burn-through. Cleaning and correct DC aluminum technique matter more than raw power.

Heat Buildup, Geometry, and the Two Faces of Melt-Through

Heat does not vanish — it accumulates. On a long seam, a clamped joint, or a small-diameter tube, each pass leaves the workpiece warmer than the last. By the final inches, the base metal can be hot enough that your starting settings are now far too aggressive. Thin wall plus small diameter means less mass to sink heat into and a shorter path for the pool to fall through. In orbital and tube work this is why schedules taper current toward the end of a rotation.

Finally, know which melt-through you are fighting. Root-pass melt-through happens at the foundation, where the full arc energy meets unsupported molten metal — usually a fit-up, purge, or travel-speed problem. Fill- or cap-pass melt-through happens later, when heat soaking through an already-thin root turns the joint into a hot sponge. Both look similar on the surface, but the fix for each is different. Build the discipline to read heat input in real time with deliberate welding practice, and melt-through becomes a variable you own — not an accident you survive.

Melt-Through by Process: TIG, MIG, Stick, Flux-Core, and Orbital

Every arc process can drop through thin metal, but the trigger and the fix change with the process. A setting that rescues a MIG root can be exactly what burns a hole in a TIG tube. Understanding the mechanics per process is how you stop chasing the defect.

TIG (GTAW): Precise and Unforgiving

TIG is where melt-through is most common on thin material. The arc is slow, focused, and only adds filler when you dip it, so on sections under about 3 mm the base metal absorbs nearly all the heat. Autogenous welds (no filler) on thin stainless and aluminum are the classic failure point: with nothing to freeze the puddle, a brief pause or a foot-pedal spike punches straight through. Polarity matters enormously — DCEN concentrates heat into the work for deep penetration, while AC supplies the cleaning action aluminum needs but splits heat between work and electrode. Our guides to TIG welding polarity and DC welding aluminium break down the trade-offs. Control comes from low amperage, fast travel, a short arc, and — on automation — downslope and pulsing.

MIG (GMAW): Where Wire Feed Meets Heat Input

MIG couples high deposition with high travel speed, so melt-through appears the moment voltage, wire feed speed (WFS), or both climb on thin sheet. Spray transfer is the worst offender: it runs hot and fluid, and on 1–2 mm steel or aluminum that runny puddle drops out before it freezes. Switching to short-circuit transfer, a smaller wire, or a pulsed mode keeps heat input inside what the metal can tolerate. A worn contact tip disturbs arc stability and can effectively raise deposition — see our contact tip welding guide.

Stick (SMAW): Thick-Plate Process, Root-Pass Risk

Stick is rarely the go-to for thin plate, so melt-through is uncommon overall. On open-root passes, though, it happens when electrode diameter or amperage is too large for the gap. Match electrode size to root opening and use a controlled, stepped root technique.

Flux-Core (FCAW): Self-Shielded vs Gas-Shielded

Self-shielded flux-core runs hotter and more aggressive than solid wire; gas-shielded flux-core behaves closer to MIG. Too much WFS on thin sheet drops through quickly. Get polarity right — most gas-shielded wires run DCEP — and match wire to thickness; our guides to polarity for flux core welding and gasless MIG welding explain the difference.

Orbital TIG: The Chronic Concern

Orbital TIG is where melt-through becomes a chronic, production-defining problem. Orbital welding covers tube diameters from 3.17 mm to 168.3 mm and wall thicknesses up to 3 mm, frequently welded autogenously, according to Orbitec. Thin-walled tubes are highly sensitive to heat input; low currents, consistent travel speeds, and controlled cooling keep distortion and penetration in check. Orbital parameter values are modified at the start of each sector because the workpiece is heated perpetually by arc energy, as the Polysoude orbital welding handbook explains. For thin-walled tubes, a torch inclined at 15° is recommended to avoid excess penetration and inner-wall melting.

SAW and High-Deposition Automation

Submerged arc and high-deposition robotic processes are built for thick plate, so on thin sections they over-thicken the weld and can drop through. High current density and deep penetration leave little margin. When these systems run near their lower thickness limits, current, travel speed, and joint preparation must be dialed back deliberately — a theme in our robotic submerged arc welding guide.

The table below summarizes risk and the first lever to pull for each process.

Simple rule: match the process, thickness, and heat input.

How to Spot Melt-Through: Visual Cues, NDT, and Root Inspection

Melt-through rarely announces itself loudly. A cap can look clean and consistent while the real damage hides on the root side, out of sight. Catching it early is a skill; catching it after the part ships is expensive. Here’s how inspectors actually find it.

Root-Side Visual Cues

The root tells the real story. Look for a convex, bulging bead that sags below the plate surface, molten “icicles” or drip tips hanging beneath the weld, and a profile that suddenly thins or opens into a ragged hole. A teardrop-shaped void or a stringer of blow-through metal behind the bead is another giveaway. Any of these means the puddle punched through.

On stainless, black or grey sugar — granular oxidation on the inside diameter — is a major red flag. Sugaring indicates lost purge gas and warns that the root had too much heat and no shielding. Discoloration in general shows how hot the back side got: a light gold tint is acceptable, but straw, blue, purple, and grey are progressively worse, and grey sugar usually means the joint is already compromised.

Cap-Side Indirect Clues

You can’t always see the root, especially inside a tube. The cap gives indirect evidence: an underfilled or concave bead, a sunken groove running down the centerline, or a section that looks suspiciously thin beside its neighbors. These hint that too much heat pushed metal through. Compare bead widths along the joint — a sudden widening often means the arc found softer, thinner metal and is about to open up.

Tools and NDT Methods

When eyes aren’t enough, bring in the right tools:

• Borescope — inspect the inside diameter of tube and pipe, including weld tube fittings, without cutting the part open.
• Weld replicas — capture the actual root profile for records and later comparison.
• Radiography (RT) — one of the main NDT methods used to detect internal and through-wall discontinuities, per the ESAB welding defects guide.
• Ultrasonic testing (UT) — finds internal flaws and loss of section thickness where RT is impractical.
• Liquid penetrant testing (PT) — locates surface-breaking cracks and discontinuities that may cause leaks in containers and pipes, as described in AWS B1.10.
• Pressure and leak testing — proves through-wall failure directly.

Short Note: PT and visual methods only find surface-breaking flaws. For buried or through-wall defects, RT, UT, or pressure testing is required.

Set up a dedicated inspection station with consistent lighting and the right eye protection; see our guides to protective eyewear for welding and welding facilities planning.

Acceptance Criteria: A Code Question, Not a Gut Call

Whether an imperfection is rejectable depends on the applicable code — not on how it looks to you. AWS D1.1, ASME, or the customer’s written specification set limits for penetration, reinforcement, and undercut. A slight convexity might pass; a through-wall hole never will. That line between cosmetic and rejectable lives in the numbers of the standard, applied by a qualified inspector.

Documentation and Weld Logs

Orbital welding is inherently documentable: programmed schedules and weld logs create an auditable record. That traceability lets you tie a rejected part back to its exact parameters — voltage, current, travel speed, and purge flow — so you can fix the cause, not just the symptom. Keep those logs alongside your inspection results, and build the visual habits that catch defects early, like those in our guide to telling good welds from bad.

Prevention Playbook: Settings, Technique, Fixturing, and Purge Control

Every welder who has blown a hole through thin metal asks the same question: how do I stop it next time? The answer is rarely one magic setting. Melt-through is a system failure, and prevention means controlling four things at once — heat input, technique, fit-up, and shielding. Get those right and thin-section welding becomes predictable instead of terrifying.

Dial In the Heat Input

Start with current. Reducing amperage and voltage lowers the total energy pouring into the joint, and shortening your arc length concentrates that energy where you actually want it. As ESAB’s welding defects guide notes, prevention begins with lower current, a tighter arc, sensible travel speed, and directing filler into the weld toes rather than the center.

For thin sheet, switch to pulse. Pulsed current alternates a hot peak with a cooler background, letting the puddle freeze between pulses. It buys you time and forgiveness. On wire processes, a worn or wrong-size contact tip causes erratic arc starts and surges — check our guide to contact tip welding for wear symptoms. Polarity matters too: wrong polarity pushes more heat into the workpiece. Review DCEN vs DCEP for flux core before you blame your hands.

Technique: Travel Speed, Angle, and Filler

Move too slowly and heat stacks up until the puddle drops through. Move well, and the arc stays ahead of the vulnerable zone. Keep a short arc, a consistent drag or push angle, and feed filler steadily into the leading edge and toes. Use the smallest filler that does the job, and prep your tungsten to a sharp point for precise, low-current starts.

Fit-Up, Fixturing, and Heat Sinking

Gaps are melt-through’s best friend. Tighten your root gap, square the edges, and clean mill scale, oil, and oxide — contamination forces you to crank heat just to get fusion. Autogenous thin-wall welds (no filler) demand precise fit-up and squareness because there is no wire to bridge a gap.

Pull heat away with copper backing bars, chill blocks, or heat sinks clamped tight to the joint. These act as thermal drains, absorbing excess energy before it burns through. Gas choice and flow also matter. Argon-rich mixes run cooler than CO2-heavy blends, and a slightly reduced flow with a tight nozzle keeps shielding without added turbulence. On aluminum, process and polarity choices change the heat balance entirely, so match the setup to the metal.

Purge Control for Tube and Pipe

On stainless and titanium tubing, internal shielding gas is not optional — it prevents oxidation and keeps the ID clean. Proper purge flow maintains a positive, stable pressure inside the tube so the root doesn’t sag or suck in atmosphere. Automatic purge control dynamically adjusts flow to hold that pressure as the joint heats and expands.

Some orbital heads go further, using a dedicated purge chamber that shields the entire welding zone so atmosphere can’t reach the seam or hot workpiece. Learn more in Polysoude’s orbital welding handbook. For manual work, tape both ends, purge fully before striking, and monitor flow with a gauge. Turbo Purge setups use high-volume, fast-fill purging to cut oxygen to near-zero before the arc starts, and tracking joint alignment keeps the gap even around the whole circumference.

Your Bench Checklist

1. Lower amperage/voltage; shorten arc length.
2. Switch to pulse for thin material.
3. Confirm polarity for your process.
4. Rebuild or replace the contact tip/tungsten.
5. Tighten root gap; square and clean edges.
6. Clamp copper backing, chill bars, or heat sinks.
7. Set purge flow; monitor with a gauge.
8. Control interpass temperature; let it cool.
9. Practice on scrap of the same thickness first.

Troubleshooting Melt-Through

None of this replaces seat time. Build the muscle memory with our welding practice guide, and treat every prevention habit as a skill you drill, not a trick you hope works.

Repairing Melt-Through Without Making It Worse

When a molten drip leaves a hole in thin material, the instinct is to strike up and fill it immediately. That instinct is usually wrong. Repairing melt-through is a controlled, methodical process — the opposite of the panic that caused it. Done right, the joint passes inspection. Done in a rush, a small hole becomes a scrapped part.

Step 1: Stop, Cool, and Assess

As soon as you see melt-through, stop the arc. Let the part cool before touching it — a heat-soaked thin section has a wide softened zone around the hole, and welding into it only spreads the damage. Once cool, assess the real extent with visual inspection, dye penetrant, or a light grind. The visible hole is almost always smaller than the damaged area beneath it.

Step 2: Excavate to Sound Metal

Defects must be excavated and re-welded with corrected parameters, with post-weld heat treatment (PWHT) where the procedure specifies it (ESAB’s welding defects guide). Use a burr or grinding wheel to remove the damaged root until you reach bright, sound metal with no undercut or oxide. Cracks are the most severe weld defects because they can propagate and cause sudden failure, so any suspected crack must be fully removed before re-welding. Watch for hidden lap defects too; our guide to the overlap welding defect covers how to tell a true weld edge from a cold, unbonded fold. If you’re working on heat-treated material, our guide on whether you can weld hardened steel explains why preheat and controlled cooling matter here.

Step 3: Close the Gap — Backing, Chill Blocks, and Copper Spoons

You can’t bridge a hole over air. Copper is a superb heat sink: a copper backing bar or “spoon” held tight behind the joint pulls heat away and supports the molten pool so you can close the opening. Clamp-on chill blocks do the same along the weld edges, while a backing strip or purge dam supports open-root work. For tubing and pipe, a gas purge on the back side does the same job as a backing strip, holding the root up while the pool freezes. Reducing the root opening with a slight re-fit also lowers the bridging load. Pair this with the right polarity for the job; our TIG welding polarity guide covers how DCEN and AC change heat balance.

Step 4: Rebuild With Lower Heat Input

Rebuild using lower current, faster travel, and a short arc. Add metal in short stitches and let each one cool before the next. The biggest trap is chasing the hole ever wider: every extra pass on thin metal heats the edges, opening them further. Keep interpass temperature low — thin sections store heat fast, and a hot part will reopen even a carefully bridged hole. If the hole grows with each attempt, stop — you’re making it worse. Keep the same WPS controls in place, because grinding out and re-welding without them just repeats the melt-through.

When to Cut It Out and Start Over

If the melt-through spans a large share of the joint, wraps a tube circumference, or sits in distorted, oxidized metal, cut it out and remake the joint. Repairing burn-through can be costly and time-consuming; the job may need to be restarted or extra weld metal added to fix the hole. A clean re-fit is usually faster and safer than endless patching.

Re-inspection and Documentation

Re-inspect with the same NDT method that caught the defect, then record the location, depth, filler, and parameters. Where the repair falls outside the qualified WPS, re-qualification is required, often against a standard such as AWS B2.1 for procedure and performance qualification. Learn the technique properly in our welding practice guide, and remember that a good repair is judged by the same standard as the original weld.

Thin Materials and Difficult Metals: Aluminum, Stainless, and Titanium

Every metal has a personality when it comes to heat. Some soak it up quietly and spread it wide; others hoard it right under the arc and melt before you notice. Melt-through lives almost entirely in the second group — the thin, the reflective, and the reactive. Understanding how each of these materials misbehaves is the difference between a clean root and a ragged hole.

Stainless Steel: The Heat Hoarder

Burn-through happens almost exclusively on thinner materials like sheet metal, and it is especially common on stainless steel because of its low heat conductivity (UNIMIG). Where carbon steel carries heat away from the puddle, stainless keeps it pinned in place. The result is distortion, warping, and — when shielding fails — sugaring, the rough, oxidized, scale-like backside that signals the purge was too weak.

Stainless also raises a health flag: its fume contains nickel and chromium, and hexavalent chromium — Cr(VI) — is regulated by OSHA under 29 CFR 1910.1026 with a permissible exposure limit of 5 µg/m³ as an 8-hour TWA.

On thin stainless, purge control is not optional. Without a steady flow of shielding gas on the backside, the hot root oxidizes as it cools, and that discoloration is a corrosion site waiting to happen. Keep travel speed steady, use the smallest practical weave, and back off amperage before you reach for a bigger puddle.

Aluminum: Fast, Reflective, and Sneaky

Aluminum is the great deceiver of thin-section welding. Its surface oxide melts near 3,700 °F — far above the roughly 1,220 °F at which the metal beneath liquefies. If that tenacious oxide layer isn’t removed, the underlying aluminum can liquefy without visible warning, creating a subsurface melt risk (UNIMIG). You see a bead forming; the material is already dropping out.

Grind oxide off with a dedicated stainless brush, then weld quickly. Aluminum conducts heat away fast, so it rewards higher amperage paired with faster travel. Choose polarity deliberately: AC helps break up the oxide, while DC is used on thicker sections — our guide to DC welding aluminium explains when each wins. And ventilate hard: NIOSH/CDC notes aluminum welding fumes can carry fluorine, arsenic, copper, silicon, and beryllium.

Titanium: Reactive and Unforgiving

Titanium reacts with oxygen, nitrogen, and hydrogen the moment it’s hot, so an airtight shield is mandatory — not just at the arc, but over the trailing puddle and the cooling bead. Extreme cleanliness, dedicated tools, and low heat input are the rules. A trailing shield or glove box keeps contamination out, and the finished color tells the story: silvery or straw is healthy, while blue, purple, or gray means oxygen slipped in. Because titanium wants DCEN (straight polarity), reviewing TIG welding polarity helps you set the machine correctly.

Galvanized and Coated Steels

Zinc melts and boils well below steel, so a galvanized surface liquefies unevenly and throws fume as you weld. Strip the coating back from the joint, increase ventilation, and watch for metal fume fever. On thin coated sheet, a gasless MIG (flux-cored) setup can help carry the arc with less fuss, though fume control still matters.

Thickness Thresholds and Joint Design

Once you drop below roughly 3 mm — thin sheet and thin-wall tube — melt-through risk climbs sharply. The fix is often mechanical, not electrical:

• Flanged edges add thickness at the joint so the arc has metal to consume.
• Lap joints give the puddle a thicker heat sink underneath.
• Backing bars (copper, ceramic, or consumable) support the root and pull heat away.
• Pulse welding drops average heat input while keeping the arc stable — a favorite for stainless and aluminum.

For any of these metals, deliberate practice with heat control pays off faster than any single setting. Our welding practice guide walks through drills that build exactly that feel.

Codes, Standards, and Acceptance Criteria: When Is Melt-Through a Defect?

The most common question about melt-through is not how to fix it — it is whether it actually matters. The answer lives entirely in the applicable code. AWS A3.0M/A3.0:2025, the 14th edition of the Standard Welding Terms and Definitions, draws the crucial line: a discontinuity is any interruption in the normal structure of a weld, while a defect is a discontinuity that makes a part unsuitable for its intended service. Melt-through is a discontinuity. Whether it becomes a defect depends on the rules you are building to.

Discontinuity or Defect? It Depends on the Code

That distinction is not semantics — it drives real money. AWS B2.1, the Specification for Welding Procedure and Performance Qualification, ties qualification to recognized references including AWS A3.0, AWS B4.0, ASME BPVC, and ANSI Z49.1. It is the framework where your WPS and PQR prove a procedure can produce sound welds, and where a welder’s performance qualification proves the individual can repeat them. Change the process, thickness, or position and the paperwork must follow. A valid PQR is what separates a defensible weld from a hopeful one.

Structural, Stainless, and Pressure Rules Differ

AWS D1.1 lets inspectors judge excess penetration against limits tied to load type, and a small amount of root reinforcement is often acceptable. AWS D1.6 applies similar logic to stainless. Pressure work flips the emphasis: ASME BPVC Section IX governs procedure and performance qualification, while ASME B31.3 governs the fabricated piping, including root profile and internal protrusion limits. Sanitary and high-purity lines — food, dairy, pharma — are the strictest of all. A crevice or rough internal bead creates a bacteria trap, so orbital and autogenous tube welds are held to flush, smooth internal profiles.

Acceptable, Limited, or Rejectable — Three Outcomes

Excess penetration is rarely a simple yes-or-no. In practice it falls into three buckets. It may be fully acceptable where the code sets no restriction on internal reinforcement. It may be acceptable with limits — a stated maximum protrusion, a length cap, or a restriction on location. Or it may be outright rejectable, such as internal protrusion that obstructs flow or creates a crevice in a hygienic system, where details like weld tube fittings demand smooth, crevice-free roots. Knowing which bucket your project uses is the whole game.

Why a Good-Looking Cap Proves Nothing

A flawless ripple finish can sit on top of a sucked-back, rejectable root. That is why deliberate welding practice includes learning to read the root, not just the cap. NDT guidance such as AWS B1.10M/B1.10:2016, the Guide for the Nondestructive Examination of Welds, catalogs these discontinuity types so you match the right test to the right flaw — visual for surface, radiography for volumetric root defects.

Controls start long before the arc. Choosing correct AC or DC for TIG welding steel keeps heat balanced and penetration predictable. In the end, the code does not care how the bead looked. It cares whether the part meets the written acceptance criteria — and that is the only judgment that counts.

A WPS is only as good as the person holding the torch. Welder performance qualification under B2.1 tests a specific individual on a specific joint, position, and process. A welder who passes on 3/16-inch plate does not automatically qualify for thin-wall tube, where melt-through risk is far higher. Matching the qualification test to the production joint is non-negotiable.

Note: When a drawing is silent on root profile, the default is to ask — not to assume.

Practical Tips, Common Mistakes, and the Bottom Line

Every melt-through you have ever produced traces back to one of six root causes: chasing a cold puddle with more amps, welding dirty metal, ignoring fit-up and gap, running thin sections with no backing, guessing at purge flow, and working without a written procedure. None of these are exotic. They are ordinary shortcuts, which is exactly why melt-through stays so common.

The cure is not a magic dial position. It is a repeatable system: know the thickness, pull parameters from a qualified procedure, control fit-up and purge, prove the setup on scrap, then inspect the root. Build that loop and melt-through stops being a surprise.

Twelve Quick-Fire Practical Tips

1. Measure material thickness with a gauge before touching a dial. “Thin” is not a measurement.
2. Set amperage from a qualified WPS or a tested chart, not from memory.
3. Control fit-up: consistent root gap, tacks every few inches, no mismatched edges.
4. Back or purge the root on thin sections—copper chill bars, purge dams, or solar flux all help.
5. Prove the setup on scrap of the same alloy and thickness before welding the real part.
6. Watch the keyhole and puddle, not the bright arc. When the puddle widens and sags, you are already late.
7. Keep travel speed steady and increase it as the part heats; heat accumulates fast on thin metal.
8. Let parts cool between passes and watch interpass temperature.
9. Clean to bright metal—mill scale, oil, and oxide all push you to add heat.
10. Hold a tight arc length and a consistent work angle; a wandering torch dumps heat.
11. Set purge flow with a meter and let it stabilize; guessing invites both porosity and burn-through.
12. Inspect every root, not just the cap. A ragged underside is a defect even when the top looks flawless.

Common Mistakes That Keep Coming Back

Chasing heat with amps is the classic error. When a puddle feels sluggish, the instinct is to crank amperage, but on thin metal that simply turns a slow weld into a hole. The real remedy is usually faster travel, a smaller wire or electrode, or better prep. Welding dirty metal is second: grease, paint, and oxide force you to add heat for fusion, and added heat is added risk. Ignoring fit-up and gap is third—a wide, uneven gap demands more filler and more energy to bridge. Skipping backing or purge leaves the root unsupported, and running without a WPS makes every weld a guess you cannot repeat or defend.

Safety Comes Before Skills

Ventilate your workspace, wear the right protective eyewear for welding, and plan fume control when you set up welding facilities.

The Bottom Line

Melt-through is a heat-balance problem you can manage. Master it and you have crossed a real milestone—the point where thin metal stops fighting you and starts obeying. If you are new to wire feed, start with How to Use a Wire Welder and build habits on scrap before chasing production. Steady procedure, clean metal, controlled purge, honest inspection. That is the whole game.

Frequently Asked Questions

What does melt-through mean in welding?

Melt-through is a condition where molten weld metal passes through the joint and appears on the root or opposite side as excess penetration, a convex bead, or even a hole. AWS and industry usage often treats it as a form of excess root reinforcement, while the full ‘drop-out hole’ is commonly called burn-through. Either way, it is a heat-input and fit-up problem.

Is melt-through the same as burn-through?

They are closely related but not identical in everyday use. Melt-through describes molten metal passing through the joint to the far side, producing excessive penetration or a sagging root. Burn-through usually refers to the extreme case where the molten pool drops out entirely and leaves a hole. Many welders use the terms interchangeably.

What causes melt-through when welding thin metal?

The main cause is too much heat for the thickness: amperage or voltage set too high, travel speed too slow, long arc length, or dwelling in one spot. Wider-than-specified root gaps, poor fit-up, lack of backing or chill bars, and dirty or coated base metal all make it worse. Aluminum and stainless steel are especially prone.

How do I prevent melt-through on thin stainless tube?

Use the lowest practical amperage, keep travel speed consistent, tighten fit-up and root gap, and use internal purge plus outside shielding gas. Chill blocks, copper backing, and pulse welding help manage heat. In orbital welding, program per-sector parameters and verify purge flow before the arc starts.

Can I repair a weld that has burned through?

Yes, but do it methodically. Stop welding, let the joint cool, assess the damage, excavate to sound metal, and rebuild with reduced heat input and correct technique. Using a backing bar or copper spoon helps bridge the gap. If the hole is large or the metal is badly thinned, cutting out and re-making the joint is often cheaper than chasing the hole.

Does polarity affect the chance of burning through?

Yes. Polarity changes how heat is distributed between the electrode and the workpiece. For example, electrode-positive (DCEP) puts more heat into the workpiece, increasing penetration and melt-through risk, while electrode-negative (DCEN) concentrates heat at the electrode and produces shallower penetration. Always follow the polarity specified in the welding procedure.

Is excess penetration always a weld defect?

Not always. Whether excess root penetration or melt-through is a rejectable defect depends on the applicable code or specification. Some structural codes tolerate limited excess penetration, while pressure, sanitary, and high-purity piping standards may reject it outright because of flow restriction, crevice formation, or corrosion concerns.

How is melt-through detected during inspection?

Visual inspection of the root side is the first check, often with a borescope inside tubing. Radiography, ultrasonic testing, and dye penetrant can detect internal or through-wall conditions. Leak and pressure tests confirm through-wall failures. In documented processes like orbital welding, weld logs and parameter records help pinpoint the cause.

Authoritative Sources & Further Reading

• https://pubs.aws.org/Download_PDFS/AWS_A3.0M_A3.0_2025_PV.pdf
• https://pubs.aws.org/Download_PDFS/b2.1PV.pdf
• https://aws-p-001-delivery.sitecorecontenthub.cloud/api/public/content/b1.10m-b1.10-2016
• https://www.osha.gov/sites/default/files/publications/OSHA_FS-3647_WELDING.pdf
• https://www.osha.gov/laws-regs/regulations/standardnumber/1926/1926.353
• https://www.cdc.gov/niosh/welding/about/index.html
• https://www.cdc.gov/niosh/npg/npgd0666.html
• https://www.sciencedirect.com/topics/materials-science/welding-defect
• 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://files.twi-global.com/technical-knowledge/faqs/faq-what-causes-crater-pipes-and-what-will-prevent-their-formation-during-welding
• https://esab.com/gb/eur_en/esab-university/articles/welding-defects-guide
• https://www.polysoude.com/wp-content/uploads/2017/01/DOC_Orbital-welding-Handbook_EN.pdf
• https://www.orbitec-group.com/en/what-is-orbital-welding