Welding troubleshooting

Why Indentation Pores Appear on the Welding Surface

Indentation pores welding surface defects are almost always a gas, wire or travel-speed problem, not a machine fault. This guide is written for engineers and shop supervisors who need to match a surface symptom to its root cause and fix it on the next pass. Work through the symptom table, then the numbered checks, and you can usually close the defect without rewriting the welding procedure.

Get Instant Quote Talk to an Engineer

Pore vs. pitGas purityWire drynessTravel speed

Indentation pores welding surface defect along a finished weld bead

On this page

Indentation pores: symptom, likely cause, action

Read the left column first. If two rows describe your part, fix the top one before touching anything else.

Symptom	Likely cause	What to do
Even row of pits along the whole bead	Gas flow too low or nozzle partly blocked	Clean nozzle, reset flow to 12–18 L/min
Pits cluster at the start of the run	Gas pre-flow too short	Add 0.3–0.5 s pre-flow before arc strike
Pits appear mid-run, then stop	Wire feed slip or spool drag	Check drive rolls, liner and spool brake
Pits near the crater only	Crater fill too fast, gas lost at the end	Slow the crater, extend post-flow to 0.5 s
Pits on one side of the bead	Torch angle or travel angle too steep	Hold 10–15° push angle, keep arc on centerline
Pits with black soot around them	Contaminated gas or moisture in the line	Purge the line, verify CO2 purity at 99.8%
Pits only on rusty or oily plate	Surface contamination boiling out	Grind back 25 mm, wipe with solvent before arc
Fine pits on every part from one spool	Wire surface dry or drawing residue	Switch spool, store wire sealed and dry

What the defect is

Indentation pores welding surface: what you are actually looking at

An indentation pore is a small depression left on the bead face after gas escapes from the molten pool. The metal was liquid, a bubble reached the surface, the bubble broke, and the pool froze before it could refill. That is why the defect reads as a pit rather than a raised blister. It is a surface symptom of something happening below the arc.

The first job is to separate an indentation pore from a pinhole and from a slag pit. A pinhole is usually round and clean, with a bright interior. A slag pit is irregular and often sits along the edge of the bead on flux-cored or stick work. An indentation pore has a smooth, rounded lip and a dull interior, because the gas left before the surface solidified.

Pores that sit in a straight line along the bead usually point to gas supply or flow. Pores that appear at random across the bead point to wire or base-metal contamination. Pores that appear only at the start or only at the crater point to timing: pre-flow, post-flow or run-in speed. Reading the pattern costs nothing and saves a lot of guesswork.

One more distinction matters for the fix. A pore that breaks the surface can be ground out and re-welded. A pore that stays just under the skin will not show until the part is machined or loaded. On parts heading to CNC finishing, we treat any visible indentation as a signal to check the whole joint, not just the spot.

Gas shielding fails in stages. A slightly low flow still covers the pool but lets air in at the edges. A badly low flow leaves the arc open to the atmosphere and produces heavy porosity. The size and spacing of the pores tells you which stage you are in, so a photograph of the bead is often enough to pick the right check.

Wire plays a quieter role. A spool that has sat in a humid shop picks up surface moisture. That moisture breaks down in the arc and adds hydrogen to the pool. The result is fine, evenly spaced pores that look nothing like the coarse pits from a gas leak. Different look, different fix.

1
Rounded lip, dull interiorGas escaped before the pool froze
2
Round and bright insideMore likely a pinhole from wire or plate
3
Lined up along the beadCheck gas flow, nozzle and hose first
4
Fine and evenly spacedSuspect wire surface moisture or drawing residue

Gas side

Gas purity, mix and flow rate on the welding surface

For CO2 welding, purity drives the result. General fabrication runs accept around 99.6%, but important joints need a minimum of 99.8%. Below that, impurities and humidity in the cylinder add oxygen and hydrogen to the arc, and both end up as pores on the welding surface. The cost difference between the two grades is small compared with a rework cycle.

Argon-rich mixes for MAG work behave the same way. Argon should be high purity, and the CO2 fraction should match the procedure, not the cylinder that happens to be on the cart. A mix that is 3–5% off its nominal CO2 content changes the arc characteristics and the pool freeze rate. Most shops never measure it. A cheap gas analyzer pays for itself in one rejected batch.

Flow rate is the second lever. Too low and the shield cannot push the atmosphere away. Too high and the gas jet turns turbulent, dragging air into the pool. For a typical 12–16 mm nozzle, 12–18 L/min covers most manual work. In a drafty bay, build a simple screen before you turn the flow up. More flow is not always more protection.

Leaks hide in the small parts. A cracked hose, a loose crimp, a worn O-ring at the regulator, or a torch body that has been dropped will pull air into the line. Leak-test with soapy water at every joint, and check the flow at the nozzle with a gauge, not at the regulator. The two readings often disagree.

Pre-flow and post-flow are timing, not flow. A short pre-flow lets the arc strike before the shield is established, which is why the first 10 mm of a bead is often the worst. A short post-flow lets the shield collapse while the crater is still liquid. Both settings cost nothing and fix a large share of start and end pores.

Moisture is the hidden variable. Compressed air lines, humid shop air and cold cylinders pulled into a warm room all put water into the gas path. If the cylinder has been stored at 5 °C and brought into a 25 °C shop, let it equalize before use. Condensation inside a regulator shows up as erratic pores that come and go with the weather.

1
CO2 purity99.8% minimum for important joints
2
Flow window12–18 L/min for a 12–16 mm nozzle
3
Pre-flow0.3–0.5 s before arc strike
4
Post-flow0.5 s or more to protect the crater

Wire side

Welding wire condition and where pores start

Solid wire is drawn from steel strip and then copper-coated. If the coating is thin or uneven, the wire surface can carry drawing residue and moisture into the arc. That is one reason two spools of the same grade can behave differently. Wipe a short length with a clean white cloth. Any dark residue on the cloth is a sign the surface is not clean.

Dryness matters as much as cleanliness. Wire stored in an open shop picks up moisture from the air. In a humid climate this happens fast. Keep spools sealed until use, keep partial spools in a dry cabinet, and never leave a spool on the machine overnight in an unheated bay. The pores from damp wire are fine and evenly spaced, and they respond well to simply changing the spool.

Alloy elements in the wire affect how the pool freezes and how it holds gas. Wire with a slightly different silicon or manganese balance can shift the pore behavior even when the grade name is the same. If a new lot causes pores and the gas settings have not changed, ask for the lot chemistry before you touch the machine. This is a common source of overnight changes in defect rate.

Feed problems mimic gas problems. A worn liner, a kinked conduit or a drive roll set too tight will make the wire slip and stall. The arc length changes, the pool gets a pause, and a pore forms. Check the liner every few spools, set drive roll tension so the wire cannot be stopped by hand, and keep the spool brake just tight enough to prevent overrun.

Consumable condition includes the contact tip and nozzle. A worn contact tip lets the wire wander, which changes arc direction and pool shape. Spatter build-up inside the nozzle disturbs the gas flow and creates turbulence. Replace the tip when the bore is visibly oval, and ream or replace the nozzle when spatter closes the opening.

Fix sequence

Step by step: clearing indentation pores from a weld

Work top to bottom. Each step takes minutes, and the order avoids chasing the wrong cause.

1
1. Record the pore patternPhotograph the bead and note whether the pores are lined up, random, at the start, or at the crater. This one minute of evidence sets the whole fix.
2
2. Verify gas flow at the nozzleAttach a flow gauge at the nozzle and set 12–18 L/min for a 12–16 mm nozzle. Compare with the regulator reading and trust the nozzle number.
3
3. Leak-test the gas pathBrush soapy water on the regulator, hose crimps and torch body. Any bubbling joint gets resealed before you weld again.
4
4. Extend pre-flow and post-flowSet pre-flow to 0.3–0.5 s and post-flow to 0.5 s or more. Run a short bead and check the start and the crater, where the defect was worst.
5
5. Check gas purity and cylinder historyConfirm CO2 at 99.8% or better for critical joints. If the cylinder was stored cold, let it equalize to shop temperature before use.
6
6. Swap the wire spoolChange to a sealed, dry spool of the same grade. If the pores stop, the old spool was damp or contaminated. Keep the old one for a lot check.
7
7. Clean the base metalGrind back 25 mm on each side of the joint and wipe with a solvent that leaves no film. Rust, mill scale and oil all boil into the pool and leave pores.
8
8. Reset travel speed and torch angleRun 10–15° push angle and a steady travel speed that keeps the pool moving. Too slow lets gas collect, too fast traps it before it can escape.

FAQs

Common questions on welding surface pores

Are indentation pores the same as porosity?

Porosity is the general term for gas cavities in a weld. Indentation pores are the ones that reach the surface and leave a visible depression. Subsurface porosity may not show at all until the part is machined or tested, which is why a surface check alone is never a full quality check.

Can I just turn up the gas flow to fix it?

Usually not. Above roughly 18 L/min on a standard nozzle, the jet becomes turbulent and pulls air into the pool, which can make pores worse. Fix leaks, nozzle condition and pre-flow before you raise the flow setting. In a drafty bay, a simple screen is a better answer than more gas.

Why do pores appear only at the start of the bead?

The shield has not built up when the arc strikes. Gas needs a moment to fill the nozzle and push the atmosphere away. A pre-flow of 0.3–0.5 s covers most manual work. If the problem persists, check for a slow solenoid or a partly blocked hose that delays gas arrival.

Does wire grade affect pore formation?

Yes, through the alloy balance and the surface condition. Silicon and manganese levels change how the pool deoxidizes and freezes. A new lot with a slightly different chemistry can shift the defect rate even when the grade name is unchanged. Ask for lot chemistry when a defect appears without a setting change.

What about base metal contamination?

Rust, mill scale, oil, paint and plating all release gas in the arc. Grind back about 25 mm on each side of the joint and wipe with a clean solvent. For galvanized or coated steel, remove the coating locally before welding, since the coating vaporizes directly into the pool.

How do I know the fix worked?

Run a test bead on scrap of the same grade and thickness, then inspect it at the same magnification you used before. Look at the start, the middle and the crater. If the pore pattern is gone in all three zones, apply the same settings to the production part and check the first piece.