Advanced Engineering Consultancy & Testing Laboratory
Centre for Advanced Testing, Inspection and Engineering Solutions
Advanced Engineering Consultancy & Testing Laboratory
Centre for Advanced Testing, Inspection and Engineering Solutions
Understand the top causes of weld cracking, from hydrogen and restraint to poor fit-up, and apply practical controls that protect structural integrity.
A crack found after a weld has cooled is rarely an isolated workmanship issue. It is evidence that the material, joint design, welding procedure, consumables, or service conditions created stresses the weldment could not accommodate. Understanding the top causes of weld cracking allows fabricators, quality managers, and asset owners to prevent defects before they become repair cycles, production delays, or integrity risks.
Weld cracking must be assessed in context. Crack location, orientation, timing, base material chemistry, weld process, heat input, restraint, and environmental exposure can each change the likely cause. The same visible crack may result from very different mechanisms, which is why an effective investigation combines inspection findings with metallurgical examination and a review of the applicable welding procedure.
Grinding out and rewelding a crack may restore the surface appearance, but it does not necessarily remove the condition that caused it. If hydrogen, excessive restraint, poor fit-up, contaminated consumables, or an unsuitable procedure remains unaddressed, cracking can recur in the repair weld or adjacent material.
For critical structural, pressure, transport, marine, energy, and mining applications, the consequences extend beyond rework cost. Cracks can reduce fatigue life, create leak paths, undermine code compliance, and initiate brittle fracture under service loading. A technically defensible response starts by identifying the crack mechanism rather than treating every crack as the same defect.
Hydrogen-assisted cracking, often called cold cracking or delayed cracking, is one of the most significant weld failure mechanisms in high-strength and hardenable steels. It commonly occurs after welding, sometimes hours or days after the joint appears acceptable. Cracks are often found in the heat-affected zone, at the weld toe, beneath the weld bead, or in the weld metal.
Three conditions generally need to coincide: diffusible hydrogen, a susceptible microstructure, and sufficient tensile stress or restraint. Hydrogen can enter through moisture in low-hydrogen electrodes or fluxes, contaminated shielding gas, damp base metal, oil, paint, rust, or poor storage practices. Rapid cooling can form hard, brittle microstructures in the heat-affected zone, particularly where carbon equivalent is elevated. Joint restraint then supplies the stress needed for cracking.
Control measures depend on the material and application, but commonly include low-hydrogen consumables maintained to manufacturer requirements, clean and dry joint faces, suitable preheat, controlled interpass temperature, heat-input control, and, where required, post-weld hydrogen diffusion treatment. A qualified procedure must account for material thickness, chemistry, restraint, and ambient conditions rather than applying a single preheat temperature to every job.
All welds shrink as they cool. When a joint is heavily restrained by its geometry, fixture arrangement, surrounding structure, or weld sequence, that shrinkage cannot occur freely. Tensile residual stresses develop and can approach the yield strength of the material.
Highly restrained joints are especially vulnerable when welding thick sections, complex assemblies, branch connections, stiffened plates, and repair areas. Even a sound filler metal can crack if the joint is forced to absorb excessive shrinkage stress. Poor assembly practices can make the problem worse: forcing misaligned components into position, using overly strong tack welds, or welding a joint with an excessive root gap can all raise restraint.
The practical response is not always to reduce heat input. Lower heat input may reduce distortion, but it can also increase cooling rate and harden susceptible heat-affected zones. Engineers should consider joint design, fit-up tolerances, balanced welding sequences, back-stepping where appropriate, temporary restraint removal, and realistic tack-weld procedures. The correct balance depends on the steel grade, section thickness, and structural configuration.
Thermal control has a direct effect on microstructure, hydrogen diffusion, and residual stress. Insufficient preheat can cause rapid cooling and increase the risk of hard heat-affected zones and hydrogen-assisted cracking. Excessive preheat or an uncontrolled interpass temperature, however, can reduce mechanical properties, increase grain growth, or alter the intended weld metal performance.
Heat input also requires control. Very low heat input can produce fast cooling and limited fusion, while very high heat input can create a wide heat-affected zone, increase distortion, and degrade toughness in some materials. The appropriate welding parameters must be demonstrated through welding procedure qualification testing and maintained in production.
Temperature measurement should be taken from the correct location using suitable instruments. Surface appearance is not a reliable indication of internal temperature, particularly on thick or thermally conductive components. Recording preheat and interpass controls provides traceability and helps identify procedural drift when cracking occurs.
Filler metal selection must match more than nominal base material strength. It must also suit the welding process, joint configuration, required toughness, service temperature, corrosion environment, and any post-weld heat treatment. A filler metal with excessive strength may create a brittle weld deposit or concentrate stress in the adjacent base metal. Conversely, insufficient strength may not meet design requirements.
Consumable condition is equally important. Low-hydrogen electrodes exposed to moisture can introduce hydrogen even when the correct classification has been selected. Flux-cored wires, submerged arc fluxes, and shielding gases must be stored, handled, and protected according to applicable requirements. Contamination from lubricants, moisture, paint, cutting fluids, or galvanizing residues can also contribute to porosity, inclusions, and cracking.
A consumable control program should include identified storage conditions, baking or reconditioning where permitted, issue-time limits, and batch traceability for critical work. These controls are often less costly than repeated repairs and destructive testing after a failure.
Cracks often begin where the welding arc is asked to compensate for poor preparation. Incorrect bevel angle, inadequate root opening, excessive root gap, high-low misalignment, unremoved backing material, and incomplete removal of thermal-cut edges can all create localized stress concentrations or poor fusion conditions.
Surface contaminants are particularly problematic. Oil, moisture, paint, oxide scale, and embedded abrasive residue may introduce hydrogen or prevent proper fusion. On thicker sections, flame-cut or plasma-cut edges can develop hard zones that require removal or treatment before welding, depending on the material and procedure.
Fit-up inspection before welding is a high-value quality control point. Confirming dimensions, cleanliness, tack weld quality, and joint accessibility helps prevent defects that are difficult to correct once multiple passes have been deposited.
Hot cracking occurs at elevated temperature while the weld metal is solidifying or shortly afterward. It is often associated with centerline cracks, crater cracks, or cracks along grain boundaries. The risk increases when the weld pool contains elements or impurities that form low-melting films at grain boundaries, when weld bead shape is unfavorable, or when shrinkage strains are high during solidification.
Material chemistry matters. Sulfur, phosphorus, and other residual elements can increase susceptibility in certain alloys. Welding parameters that create a deep, narrow bead profile may also promote centerline solidification cracking. In aluminum, stainless steels, nickel alloys, and some high-alloy materials, filler selection and dilution control are particularly relevant.
Preventive measures may include selecting a compatible filler metal, adjusting travel speed and current to improve bead profile, controlling dilution, filling weld craters properly, and maintaining clean base material. Because hot cracking is alloy-specific, procedure qualification should reflect the actual material combination and production welding conditions.
Lamellar tearing occurs in rolled plate when weld shrinkage imposes tensile strain through the plate thickness. It generally develops beneath or adjacent to welds in highly restrained T-joints, corner joints, and attachments. The crack path may step through the plate parallel to the rolling plane, often following elongated nonmetallic inclusions.
This mechanism is not solved solely by changing welding parameters. Joint design and steel quality are central factors. Solutions may include using plate with improved through-thickness properties, revising the joint to reduce through-thickness strain, applying buttering layers where appropriate, or changing the weld sequence and attachment arrangement.
Some cracks develop after fabrication rather than during welding. Reheat cracking can occur during post-weld heat treatment or elevated-temperature service in susceptible alloy steels. Fatigue cracking may initiate at weld toes, undercut, lack-of-fusion locations, or geometric transitions under cyclic loading. Corrosion-assisted cracking can occur where tensile stress combines with a damaging environment.
These mechanisms require a broader asset-integrity review. The original weld may satisfy visual acceptance criteria but still contain a stress concentration or metallurgical condition that becomes significant in service. Loading history, operating temperature, environment, coating condition, and prior repairs should be reviewed alongside the weld records.
When cracking is discovered, the first priority is to preserve evidence. Record crack location, dimensions, orientation, weld identification, joint details, and the time at which the indication was found. Avoid extensive grinding before inspection, as it can remove features that distinguish a crater crack from a heat-affected-zone crack or a fatigue-related indication.
A suitable investigation may combine visual testing, magnetic particle testing or liquid penetrant testing, ultrasonic testing, radiography where applicable, hardness mapping, chemical analysis, and metallographic examination. Fractography using stereomicroscopy or scanning electron microscopy can be valuable where the failure mechanism is uncertain. Review welding parameters, consumable records, preheat logs, material certificates, fit-up records, and any heat-treatment data alongside the examination results.
AECTL can support this process through accredited materials testing, weld inspection, failure analysis, and engineering consultancy, helping project teams move from a visible defect to a defensible root-cause assessment and corrective action.
The most effective crack prevention program is built before the first arc is struck: verify the material, qualify the procedure for the real joint conditions, control consumables and temperatures, inspect fit-up, and treat unexpected cracking as engineering evidence. That approach protects both the weld and the decisions that depend on it.