Contamination, Porosity, Oxidation, and Cracking: Diagnosing and Fixing Common TIG Problems

How to Use This Defect Reference

When a TIG weld suddenly looks wrong, treat it like troubleshooting a process, not “bad luck.” Each defect below is organized the same way: what it looks like, likely causes, quick tests, and corrective actions. Many problems share the same contamination pathways: dirty base metal, oxidized filler, inadequate post-flow, drafts, and excessive arc length. The fastest way back to clean welds is to identify which pathway is active, then restore clean conditions with a short, repeatable reset.

Contamination Pathways (the usual suspects)

• Dirty base metal: oils, paint, marker ink, cutting fluid, grinding dust, rust/scale, fingerprints.
• Filler oxidation/contamination: rod stored uncovered, touched with bare hands, dipped into a dirty puddle, end oxidized from heat.
• Shielding failure: inadequate post-flow, drafts/fans, loose fittings, damaged hoses, wrong cup size, poor coverage at corners.
• Technique-driven contamination: excessive arc length, wandering torch angle, overheating the tungsten, dipping tungsten into the puddle, dragging filler through the gas boundary.

Porosity (Pinholes, Worm Tracks, “Bubbly” Bead)

What it looks like

• Small round pinholes on the surface or revealed after brushing/grinding.
• “Worm tracks” (elongated pores) along the bead.
• Puddle looks like it is fizzing or releasing tiny bubbles.

Likely causes

• Contaminants on base metal (oil, solvent residue, paint, plating, cutting fluid).
• Moisture: damp filler, condensation on cold metal, wet gloves.
• Shielding gas disruption: drafts, too much/too little flow for the setup, poor cup coverage, torch too far from work.
• Excessive arc length pulling air into the shielding envelope.
• Leaky gas path: loose back cap, cracked torch body, pinched hose, leaking regulator connection.

Quick tests

• Draft test: hold a strip of tissue near the arc zone (away from heat) to see airflow movement; if it flutters, shielding is at risk.
• Gas integrity check: listen/feel for leaks at connections; spray soapy water on fittings (with gas flowing) and look for bubbles.
• Arc-length check: run a short bead while consciously shortening arc length; if porosity improves immediately, technique and shielding envelope were the issue.
• Filler-end check: inspect the last 25–50 mm of rod; if it’s dark/oxidized, cut it off and retry.

Corrective actions

• Reclean the joint: remove coatings/scale mechanically, then wipe with a clean, lint-free cloth and appropriate solvent; let it fully flash off before welding.
• Refresh filler: cut off oxidized rod end; wipe rod; store rods in a clean tube or bag.
• Restore shielding coverage: increase cup size or use a gas lens when needed; set flow appropriate to cup and environment; keep torch closer and steadier.
• Eliminate drafts: block airflow, reposition work, turn off fans aimed at the weld zone.
• Shorten arc length: keep the tungsten close enough to maintain a tight, stable arc and a compact shielding envelope.

Gray/Black Oxidation (Dull, Sooty, Sugary-Looking Surface)

What it looks like

• Bead and heat-affected zone turn dull gray, dark gray, or black.
• Surface may look chalky, crusty, or sooty; edges may appear “burned.”
• Color worsens at starts/stops or when changing direction.

Likely causes

• Inadequate shielding during welding (drafts, poor coverage, wrong cup position).
• Inadequate post-flow allowing the hot tungsten and crater to oxidize.
• Excessive arc length and/or torch angle that exposes the puddle to air.
• Gas flow problems: too low (air intrusion) or too high (turbulence pulling air in), or a partially blocked cup screen/lens.
• Contaminated base metal or filler contributing to dirty surface appearance.

Quick tests

• Post-flow test: increase post-flow and watch the tungsten color after stopping; if it stays bright longer and oxidation reduces, post-flow was insufficient.
• Flow/turbulence test: reduce flow slightly if you suspect turbulence (noisy, harsh gas stream) and compare bead color.
• Cup distance test: run a short bead with the cup closer and more centered; improved color indicates coverage/angle issues.

Corrective actions

• Increase post-flow so shielding continues until the tungsten and crater cool below oxidation-prone temperature.
• Stabilize torch position: keep a consistent, modest torch angle and a short arc length; avoid “reaching” into corners with the tungsten far out.
• Optimize gas delivery: verify cup is clean, lens screens are not clogged, and flow matches the cup size and environment.
• Reclean and restart: if oxidation is heavy, stop, clean the area back to bright metal, and restart with corrected shielding.

Tungsten Inclusions (Hard Specks, Streaks, or “Pepper” in the Bead)

What it looks like

• Bright metallic flecks or dark specks embedded in the bead.
• Small raised bumps that won’t brush off.
• Arc becomes unstable; bead may show random contamination trails.

Likely causes

• Tungsten dipped into the puddle or touched filler.
• Overheated tungsten from excessive amperage for the tip size, too long stick-out, or poor gas coverage.
• Contaminated tungsten from previous contact with base metal, filler, or dirty grinding wheel.
• Arc wandering due to long arc length, magnetic arc blow (DC), or poor work clamp contact.

Quick tests

• Tip inspection: stop and look at the tungsten; any blob, discoloration, or uneven tip indicates contamination.
• Scratch test on a clean plate: if the arc is unstable or “spitty” on clean material, the tungsten is likely compromised.
• Stick-out check: reduce tungsten stick-out and see if stability improves.

Corrective actions

• Regrind or replace tungsten: remove contaminated portion fully; grind on a dedicated, clean wheel/belt; keep the tip geometry consistent.
• Reduce dip risk: slightly shorten arc length, slow down filler hand, and keep filler within the shielding zone before it reaches the puddle.
• Match tungsten size to amperage: if the tip overheats repeatedly, increase tungsten diameter or reduce current/duty cycle.
• Improve coverage: use a larger cup or lens and reduce stick-out to protect the hot tip.

Lack of Fusion (Bead Sitting on Top, Cold Lap, Unbonded Sidewall)

What it looks like

• Bead has a rounded “rope” appearance and doesn’t blend into the toes.
• Visible line at the toe where bead meets base metal.
• On fillets, one leg looks stuck on without wetting; on butts, the edges don’t tie in.

Likely causes

• Insufficient heat at the joint: travel too fast, amperage too low for thickness, or heat not directed into the sidewall/root.
• Arc length too long reducing heat density and control.
• Incorrect torch aim: heating filler more than base metal, or favoring one side unintentionally.
• Oxide/scale acting as a barrier (especially if the joint faces weren’t brought to bright metal).
• Overfeeding filler chilling the puddle and preventing wetting.

Quick tests

• Pause test: briefly pause at the toe/sidewall; if it suddenly wets in, you were outrunning fusion.
• Amperage bump: increase current slightly for a short section; if tie-in improves, heat input was marginal.
• Arc-length discipline: run a bead focusing only on short arc length; improved wetting indicates arc length was the main limiter.

Corrective actions

• Re-aim the arc: point the arc at the thicker member or the sidewall/root you need to fuse, then add filler into the leading edge of the puddle.
• Adjust heat input: increase amperage, slow travel slightly, or reduce filler addition rate so the puddle can wet in.
• Clean to bright metal: remove scale/oxide at the toes and joint faces; contamination can mimic “cold” technique.
• Maintain short arc length: tighter arc improves penetration and directional control.

Undercut (Groove Along the Toe of the Weld)

What it looks like

• A visible channel melted into the base metal along one or both toes.
• Toe looks sharp and thin instead of smoothly blended.
• Often appears on outside corners, thin material, or when moving fast.

Likely causes

• Too much heat for the travel speed and joint geometry.
• Travel speed too fast while maintaining high amperage, leaving no time for the puddle to fill the edges.
• Arc too long spreading heat and washing the edges away.
• Not enough filler to support the toes (especially on fillets).
• Torch angle pushing the puddle off the edge rather than letting it fill.

Quick tests

• Slow-down test: reduce travel speed slightly while adding a touch more filler; if undercut disappears, you were outrunning fill.
• Arc-length test: shorten arc length; if the groove reduces, arc spread was contributing.
• Heat step-down: reduce amperage a small amount and compare toe shape.

Corrective actions

• Balance heat and fill: lower amperage or slow down and add filler so the toes are supported.
• Refine torch angle: keep the arc centered and avoid pushing molten metal away from the edge.
• Use controlled pauses: slight dwell at each toe (micro-pause) can help fill without overheating the center.
• Repair approach: lightly dress the undercut area to clean metal and re-run a controlled pass to blend the toe.

Crater Cracking (Cracks at the End of the Weld)

What it looks like

• Small star-shaped or straight crack at the weld termination.
• Crater is concave (a “divot”) with a thin center.
• Crack may be visible immediately or after cooling.

Likely causes

• Stopping abruptly without filling the crater.
• High restraint in the joint (tacks, clamps, rigid fit-up) concentrating shrinkage stress at the end.
• Overheating near the end, leaving a large, fluid crater that shrinks hard.
• Inadequate post-flow can worsen end-of-weld oxidation and weaken the crater area.

Quick tests

• Termination technique test: on a practice piece, taper current down slowly while adding a final small amount of filler; if cracks stop, termination was the issue.
• Crater shape check: if you consistently leave a divot, you’re setting up a crack starter.

Corrective actions (step-by-step)

1. Plan the stop: as you approach the end, reduce travel speed slightly to stabilize the puddle.
2. Add filler to fill the crater: deposit a small amount so the end is slightly crowned, not concave.
3. Taper off heat: use a controlled downslope or foot pedal roll-off while holding the torch steady.
4. Hold shielding: keep the torch in place during post-flow to protect the hot crater and tungsten.

Warping/Distortion (Parts Pulling, Bowing, Misalignment)

What it looks like

• Panels curl, angles close/open, butt joints pull out of alignment.
• Gaps change as you weld; tacks feel like they “fight” the fit-up.
• Finished assembly rocks on a flat surface.

Likely causes

• Excessive heat input: long dwell, high amperage, wide beads, repeated passes without cooling.
• Unbalanced welding sequence: welding continuously from one end to the other on thin sections.
• Insufficient fixturing or poor tack strategy for the part’s stiffness.
• Overbuilding the bead: more weld metal than needed increases shrinkage.

Quick tests

• Heat audit: note if distortion increases after long continuous beads; if yes, reduce continuous length and allow cooling.
• Sequence test: try short, alternating stitch segments on scrap; if distortion reduces, sequencing was the driver.

Corrective actions

• Reduce heat input: smaller bead, faster travel with adequate fusion, and avoid unnecessary dwell.
• Use balanced sequencing: alternate sides, skip around, or backstep in short sections to distribute shrinkage.
• Improve restraint smartly: clamp to a heat sink or fixture, but avoid over-restraint that can promote cracking in some alloys.
• Let it cool: controlled pauses between passes can prevent cumulative heat buildup.

Fast “Restore Clean Conditions” Protocols

Recleaning protocol (when contamination is suspected)

1. Stop welding as soon as the puddle looks dirty or the arc becomes unstable.
2. Remove the bad section: grind or file back to sound metal; don’t weld over porous/oxidized material.
3. Mechanical clean: dedicated abrasive for the material; remove scale/oxide and any heat tint where needed.
4. Solvent wipe: lint-free cloth; wipe in one direction; allow full evaporation.
5. Protect the cleaned area: avoid touching with bare hands; keep filler and tools off the joint faces.

Regrinding protocol (when tungsten contamination is suspected)

1. Power down/secure torch and remove tungsten.
2. Cut back to remove any contaminated/balling area (don’t just “touch up” the very tip if contamination traveled).
3. Grind on a clean, dedicated abrasive and keep the grind consistent; avoid embedding foreign material.
4. Reassemble and verify: correct stick-out, tight back cap, clean collet, clean cup/lens.

Gas coverage quick-corrections

• If oxidation appears suddenly: check for drafts first, then verify flow, then inspect cup/lens for blockage or damage.
• If porosity appears intermittently: suspect loose fittings, cracked hose, or turbulence from excessive flow.
• If starts/stops are dirty: increase pre/post-flow and keep the torch in place until post-flow ends.

Repeatable Reset Procedure (When Weld Quality Suddenly Degrades)

Use this sequence in order; it’s designed to restore a clean, stable process quickly without guessing.

1. Freeze and observe: note whether the problem is in the puddle (bubbling/dirty), on the surface (gray/black), or in arc behavior (wandering/spitting).
2. Check shielding environment: eliminate drafts; reposition the work so the cup can cover the joint; verify torch is not too far from the puddle.
3. Verify gas delivery: confirm gas is on, flow is reasonable for the cup, and fittings are tight; inspect cup/lens screens for spatter/dirt; ensure back cap and torch parts are snug.
4. Inspect tungsten: if the arc is unstable or the bead shows specks, stop and regrind/replace tungsten using the regrinding protocol.
5. Refresh filler: cut off the heated/oxidized end; wipe the rod; keep it inside the shielding zone during feeding.
6. Reclean the joint: grind back defects to sound metal; mechanically clean; solvent wipe; let dry fully; avoid touching.
7. Re-run a short test bead: 25–50 mm on clean scrap or a clean section of the part; focus on short arc length and steady torch angle.
8. Only then adjust technique/heat: if the bead is clean but fusion/shape is wrong, make small changes (travel speed, filler rate, heat input) one at a time and retest.