3.3 Porosity and Inclusions

Key Takeaways

  • Porosity is gas trapped during solidification; inclusions are trapped solids — both are volumetric
  • Four porosity types: uniformly scattered, cluster, linear, and piping (wormholes)
  • Wind is the #1 outdoor GMAW porosity cause; inadequate interpass cleaning is the #1 slag-inclusion cause
  • Tungsten inclusions occur in GTAW when the electrode touches the pool; they read bright on radiographs
  • AWS D1.1: CJP butt welds transverse to tensile stress permit NO visible piping porosity
Last updated: June 2026

Volumetric Discontinuities

Porosity and inclusions are volumetric (three-dimensional, rounded) discontinuities — they occupy volume inside the weld rather than presenting a sharp planar face. Because their notch is blunt, they are generally less damaging than cracks or lack of fusion for the same size, but they are routinely the most frequently encountered discontinuities and will reject a weld once they exceed the code's size, frequency, or distribution limits. The CWI exam tests you on types, causes, and the most likely corrective action for a described situation.

Porosity

Per AWS A3.0, porosity is cavity-type discontinuities formed by gas entrapment during solidification. Dissolved gas (most commonly hydrogen, also nitrogen, oxygen, CO) tries to escape the freezing pool; if it is trapped before solidification, it leaves a rounded void.

Types of Porosity

TypeAppearanceTypical cause
Uniformly scatteredSmall pores evenly distributed through the weldGeneral mild contamination of base metal or consumable
Cluster (localized) porosityA group of pores in one spotLocalized contamination, poor arc starts/stops
Linear (aligned) porosityPores in a rowContamination along a specific interface — root, tack weld, interpass line
Piping porosity (wormholes)Elongated tubular pores, axis roughly perpendicular to the weld faceGas (often from base-metal coatings/galvanizing or moisture) evolving and channeling during solidification

Causes of Porosity — Gas Source

CauseGas generated
Moisture on plate, electrode coating, or fluxHydrogen / water vapor
Oil, grease, paint on the jointHydrogen, carbon compounds
Rust or mill scaleOxygen, carbon monoxide
Inadequate shielding-gas coverage / low flowNitrogen, oxygen from air
Wind or drafts disturbing the gas shieldAtmospheric N₂/O₂
Excessive arc length (SMAW/GMAW)Air entrainment into the arc
Damp low-hydrogen flux/electrode (improper storage)Hydrogen
Welding over porous tack weldsTrapped gas from the tack

"* Map the clue to the gas source. ** Damp electrodes left out of the oven → hydrogen. Rust/scale not cleaned → oxygen/CO. Long arc → air entrainment. Porosity may be surface-breaking (visible during VT) or fully subsurface (found only by RT, where pores read as dark, well-rounded spots). Because the gas pocket is rounded, porosity is one of the more tolerable discontinuities — but clustered, linear, or piping porosity concentrated at the root or transverse to tension is treated far more strictly than the same volume of scattered pores.

Inclusions

An inclusion is solid foreign material entrapped in the weld metal or between passes — distinct from porosity, which is a gas cavity.

InclusionDescriptionPrimary cause
Slag inclusionNonmetallic solid (slag from a flux/coating process: SMAW, FCAW, SAW) trapped in the weldInadequate interpass slag removal, too-fast travel letting slag run ahead of the puddle, undercut pockets, poor bead profile/sequence
Tungsten inclusionParticle of tungsten electrode embedded in the weld (GTAW only)Electrode dipped into the molten pool, or too much current for the tungsten diameter; shows as a bright spot on a radiograph (tungsten is dense)
Oxide inclusionSurface oxide (e.g., aluminum oxide) folded into the weldInadequate shielding or surface cleaning, especially on aluminum

Slag-inclusion prevention is a recurring exam point: remove all slag between every pass (chip, wire-brush, or grind); use a proper bead sequence so you don't create sharp valleys that trap slag; keep travel speed correct so slag does not solidify ahead of the molten metal; and manipulate the arc to keep slag floating behind the puddle. Tungsten-inclusion prevention: keep the tungsten out of the pool (proper arc length and technique) and size the electrode to the amperage.

Acceptance Criteria — AWS D1.1 Clause 8

Visual and volumetric limits differ for statically versus cyclically loaded connections, with cyclic (fatigue) members held to tighter limits.

Condition (per AWS D1.1)Limit
CJP groove welds in butt joints transverse to computed tensile stressNo visible piping porosity
Other groove welds and fillet weldsSum of visible piping porosity ≥ 1/32 in. diameter ≤ 3/8 in. in any linear inch, and ≤ 3/4 in. in any 12 in. of weld
Fillet welds — piping porosity frequencyone in each 4 in. of length; max diameter ≤ 3/32 in.

Subsurface slag and porosity in radiographically or ultrasonically tested welds are judged against Clause 8 RT/UT tables based on size and spacing. The big idea: rounded, well-distributed porosity is tolerated within limits; concentrated, aligned, or transverse-tension piping porosity is treated most harshly.

Exam essentials: Porosity = trapped gas; inclusions = trapped solid. Wind is the #1 outdoor GMAW porosity cause; inadequate interpass cleaning is the #1 slag cause; tungsten in the pool causes tungsten inclusions (GTAW). Know that CJP butt welds transverse to tension permit no visible piping porosity.

Porosity Types, Inclusion Types, and D1.1 Numbers

The CWI must name each porosity pattern, map it to a gas source, and recall the D1.1 acceptance numbers for piping porosity. Pulling the key facts together:

Porosity Patterns and Gas Sources

PatternLookGas source clue
Uniformly scatteredSmall pores spread through the weldGeneral contamination of base metal or consumable
Cluster (localized)A group of pores in one spotPoor arc starts/stops, localized contamination
Linear (aligned)Pores in a row along an interfaceContamination at the root, a tack, or an interpass line
Piping (wormhole)Elongated tubular pores, axis roughly perpendicular to the weld faceGas from base-metal coatings, galvanizing, or moisture channeling during freezing

The exam often gives a scenario and asks for the cause: damp electrodes out of the oven → hydrogen; uncleaned rust or mill scaleoxygen / carbon monoxide; wind or excessive arc length → atmospheric nitrogen / oxygen entrained into the arc. Porosity may be surface-breaking (caught in VT) or fully subsurface (read on RT as dark, well-rounded spots).

Inclusion Types

  • Slag inclusions — nonmetallic solid from a flux process (SMAW, FCAW, SAW) trapped by inadequate interpass cleaning, too-fast travel letting slag run ahead of the puddle, or sharp bead valleys. The #1 slag cause is inadequate interpass slag removal.
  • Tungsten inclusions — a particle of the GTAW electrode dropped into the pool, or too much current for the tungsten diameter; on a radiograph tungsten is dense, so it reads as a bright white spot (most other discontinuities read dark).
  • Oxide inclusions — surface oxide folded into the weld, common on aluminum with poor cleaning or shielding.

AWS D1.1 Acceptance for Piping Porosity

ConditionLimit
CJP butt joints transverse to computed tensile stressNo visible piping porosity
Other groove and fillet weldsSum of piping porosity ≥ 1/32 in. ≤ 3/8 in. per linear inch and ≤ 3/4 in. per 12 in.
Fillet welds — frequencyone in each 4 in. of length; max diameter ≤ 3/32 in.

The governing idea: rounded, well-distributed porosity is tolerated within limits, but concentrated, aligned, or transverse-tension piping porosity is judged most harshly — and a CJP butt weld transverse to tension permits no visible piping porosity at all.

Test Your Knowledge

A welder running GMAW outdoors produces uniformly scattered porosity. What is the MOST likely cause?

A
B
C
D
Test Your Knowledge

Which discontinuity is caused specifically by a GTAW electrode contacting the molten weld pool?

A
B
C
D
Test Your Knowledge

Per AWS D1.1, a CJP groove weld in a butt joint transverse to the computed tensile stress is allowed how much visible piping porosity?

A
B
C
D