3. Welding Metallurgy and Weldability

Introduction and Context

Welding is not merely a joining operation; it is a localized thermal and metallurgical treatment of engineering materials. During welding, a heat source melts a region of the base metal and, in many processes, a filler metal. The molten pool subsequently solidifies, while adjacent material experiences thermal cycles below its melting temperature. Consequently, a completed weld is a heterogeneous system consisting of weld metal, fusion boundary, heat-affected zone (HAZ), and unaffected base metal. Each region may possess different microstructure, composition, hardness, residual stress, and mechanical properties.

Welding metallurgy provides the scientific basis for predicting these changes. It explains how heat flow controls solidification, how alloying elements partition between phases, how grains grow, and why defects such as porosity, hot cracking, hydrogen cracking, and lack of toughness occur. Weldability is the practical expression of this knowledge: it is the ability of a material to be welded under specified fabrication conditions so that the resulting joint meets required service performance.

For postgraduate engineers, weldability cannot be judged from composition alone. Joint design, restraint, heat input, thermal cycle, shielding, consumable selection, preheating, interpass temperature, and post-weld heat treatment must be considered together. Metallurgical reasoning therefore connects welding procedure qualification, inspection, failure analysis, and process selection.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Describe heat flow and solidification in fusion welding.

2. Explain phase transformations and grain growth in weld metal and HAZs.

3. Interpret segregation, dilution, and their effect on joint properties.

4. Distinguish the major HAZ subregions and relate them to thermal cycles.

5. Calculate and interpret carbon equivalent as an indicator of hardenability and hydrogen-cracking risk.

6. Explain hot cracking, cold cracking, lamellar tearing, and other metallurgical failure mechanisms.

7. Select appropriate weldability tests and interpret their limitations.

8. Apply metallurgical principles to procedure development and industrial problem-solving.

The Weld as a Thermally and Chemically Heterogeneous System

A fusion weld is commonly divided into four regions. The weld metal is formed from the solidified molten pool and may contain both filler metal and melted base metal. The fusion boundary is the narrow interface at which liquid metal has solidified against partially melted or unmelted base metal. The HAZ is the base-metal region whose microstructure or properties have been altered by heat, although it has not melted. Beyond the HAZ lies the unaffected base metal.

The thermal cycle at a point is defined by its heating rate, peak temperature, time above critical temperatures, and cooling rate. Two points at the same distance from a weld may experience different cycles when welding speed, plate thickness, joint geometry, or heat dissipation changes. A thermal cycle determines transformation products, grain size, precipitation behavior, residual stress, and hardness.

Heat input is often estimated for an arc process by

H = ηVI/v,

where H is effective heat input per unit length, η is thermal efficiency, V is voltage, I is current, and v is travel speed. In practical calculations, consistent units are essential. A higher heat input generally enlarges the HAZ and reduces cooling rate, although the actual effect depends strongly on thickness, joint geometry, preheat, and heat sinking. Heat input is therefore a useful control variable but not a complete description of a weld thermal cycle.

Solidification of Weld Metal

Solidification begins when the liquid weld pool loses sufficient heat to the surrounding solid metal. Nucleation is usually heterogeneous and occurs preferentially at the fusion boundary. Existing grains in the base metal can grow epitaxially into the liquid because the crystal lattice at the interface provides favorable nuclei. The growth direction is controlled by the maximum temperature gradient, and columnar grains commonly extend toward the centre of the weld pool.

The morphology of solidification depends on the ratio of temperature gradient, G, to interface growth rate, R. A high G/R ratio favors planar or cellular growth, whereas lower values promote cellular-dendritic or equiaxed structures. The centre of a weld may contain equiaxed grains when constitutional undercooling allows new grains to form in the liquid. Grain morphology affects anisotropy, toughness, segregation, and crack susceptibility.

Solidification is also influenced by alloy constitution. In steels, weld-metal transformations may produce ferrite, pearlite, bainite, acicular ferrite, or martensite. In stainless steels and nickel alloys, austenite-ferrite balance, carbide formation, and low-melting eutectics are important. A weld-metal composition designed only to match nominal base-metal strength may be unsuitable if it produces excessive hardenability or a crack-sensitive solidification structure.

Segregation During Solidification

Most alloying elements have different solubilities in liquid and solid phases. If an element is rejected by the growing solid, it becomes enriched in the remaining liquid. This microsegregation is greatest between dendrite arms and may produce low-melting films or brittle constituent particles. Macrosegregation can occur when fluid flow, buoyancy, or shrinkage transports enriched liquid over larger distances.

Segregation is significant because crack initiation often occurs in the final liquid to solidify. Sulfur and phosphorus are especially harmful in many steels, while niobium, molybdenum, boron, and other alloying elements may influence precipitation and liquation. Reducing impurity levels, controlling filler composition, refining weld-metal chemistry, and optimizing heat input are common measures for limiting segregation-related problems.

Phase Transformations and Grain Growth

The HAZ experiences a sequence of transformations determined by peak temperature. In carbon and low-alloy steels, the coarse-grained HAZ lies immediately next to the fusion boundary and is heated well above the austenitizing range. Austenite grains grow rapidly in this region, and subsequent cooling may produce hard bainite or martensite. The coarse-grained HAZ is often the least tough part of a welded joint.

The fine-grained HAZ is heated sufficiently to transform to austenite but at a lower peak temperature than the coarse-grained region. New grains nucleate more readily, producing a refined structure that may have better toughness. The intercritical HAZ is heated between the lower and upper critical temperatures. Only part of the original microstructure transforms, and local enrichment or untempered martensite may develop during cooling. The subcritical HAZ experiences tempering, recovery, precipitation, or softening without complete austenitization.

In quenched-and-tempered steels, the HAZ may show either hardening or softening. Hardening results when rapid cooling forms martensite; softening occurs when pre-existing tempered martensite is overtempered. In precipitation-hardened aluminium alloys, welding can dissolve or coarsen strengthening precipitates, creating a softened HAZ. In austenitic stainless steels, sensitization may occur if chromium carbides precipitate at grain boundaries within a critical temperature range, reducing resistance to intergranular corrosion.

Grain growth is promoted by high peak temperature and prolonged exposure. It reduces grain-boundary area and can lower toughness. Grain growth can be limited by controlling heat input, using multiple low-heat-input passes where appropriate, employing controlled thermal cycles, and selecting alloys or consumables with suitable grain-refining constituents. However, excessively low heat input may increase cooling rate and promote hard, brittle transformation products. Metallurgical control therefore requires balancing competing effects.

Dilution and Weld-Metal Chemistry

Dilution is the proportion of weld metal originating from the base metal rather than the filler. In simplified form,

Dilution (%) = base-metal contribution / total deposited weld metal × 100.

Dilution depends on process, polarity, current, travel speed, joint geometry, and deposition technique. It is especially important when a filler metal is selected to provide corrosion resistance, wear resistance, or a controlled phase balance. Excessive dilution can lower alloy content and change ferrite number, hardness, corrosion resistance, or crack susceptibility.

For dissimilar-metal joints, dilution may create a composition that lies in an undesirable phase field. Buttering, buffer layers, controlled filler selection, or modified welding parameters may be used to manage this problem. Chemical analysis of deposited metal and metallographic examination are useful when dilution is critical.

Carbon Equivalent and Hardenability

Carbon equivalent (CE) expresses the combined effect of carbon and alloying elements on hardenability relative to carbon. A widely used form for many low-alloy steels is the IIW equation:

CE(IIW) = C + Mn/6 + (Cr + Mo + V)/5 + (Ni + Cu)/15,

where alloy contents are in weight percent. Other formulas, including the carbon equivalent or crack-susceptibility index used in specific standards, may be more suitable for low-carbon modern steels. The formula must therefore be identified before a value is interpreted.

A high CE generally indicates a greater tendency to form hard martensite in the HAZ and a higher risk of hydrogen-assisted cold cracking. It does not directly predict cracking by itself. Plate thickness, restraint, diffusible hydrogen, preheat, cooling time, joint geometry, and residual stress are also decisive. Preheating lowers cooling rate, allows hydrogen to diffuse away, and reduces the formation of untempered martensite. Low-hydrogen consumables and controlled storage are equally important.

The cooling time from 800 to 500°C, commonly expressed as t8/5, is a practical parameter in steel welding. A short t8/5 may produce high hardness, while an excessively long t8/5 may cause coarse grains or undesirable softening. Procedure qualification often seeks a controlled range rather than simply maximizing or minimizing heat input.

Cracking Mechanisms in Welded Joints

Hot or Solidification Cracking

Hot cracking occurs during solidification or at high temperature when the material has insufficient ductility to accommodate contraction. It is promoted by a wide freezing range, low-melting segregated films, high sulfur or phosphorus, unfavourable weld geometry, and tensile strain. Centreline cracking is common in highly restrained welds with deep, narrow profiles. Prevention includes cleaner base materials, suitable filler chemistry, optimized bead shape, reduced restraint where possible, and controlled welding parameters.

Liquation cracking occurs in the partially melted zone or HAZ when constituents at grain boundaries melt locally and are unable to withstand contraction. It is particularly relevant to some nickel alloys, aluminium alloys, precipitation-hardened steels, and high-alloy materials.

Hydrogen-Assisted Cold Cracking

Hydrogen-assisted cracking usually develops after cooling, often in the HAZ, and requires three conditions: diffusible hydrogen, a susceptible microstructure, and tensile stress. Hydrogen enters through moisture, contamination, damp electrodes, fluxes, shielding gases, or surface oxides. It diffuses toward traps and highly stressed regions. Hard martensite and high restraint increase susceptibility.

Control is based on a coordinated strategy: use low-hydrogen consumables, clean and dry joint surfaces, apply specified preheat, control interpass temperature, maintain adequate heat input, reduce restraint where practicable, and apply post-weld hydrogen diffusion or heat treatment when required. Delayed inspection is sometimes specified because cracking may not appear immediately.

Lamellar Tearing

Lamellar tearing is a step-like fracture in the base metal, generally parallel to the plate surface, caused by through-thickness strains acting on non-metallic inclusions. It is associated with highly restrained T-joints, cruciform joints, and thick plates having poor through-thickness ductility. Prevention may involve improved through-thickness material, joint redesign, buttering, smaller welds, or altered welding sequence.

Reheat Cracking and Other Failures

Reheat cracking can occur in susceptible creep-resistant steels during post-weld heat treatment or service at elevated temperature. It is associated with precipitation processes and stress concentration at coarse-grained HAZ boundaries. Underbead cracking, crater cracking, and stress-corrosion cracking may involve combinations of metallurgical and mechanical factors. Diagnosis should therefore consider fracture location, morphology, thermal history, chemical composition, and service environment rather than relying on a single indicator.

Weldability Testing and Evaluation

No single test represents all service conditions. Weldability assessment should combine chemical analysis, thermal-cycle information, metallography, hardness measurement, mechanical testing, and nondestructive examination.

The restraint cracking test evaluates susceptibility under controlled restraint and may be designed for longitudinal or transverse cracking. The Tekken (Y-groove) test is widely used for hydrogen-assisted cracking in steels; it imposes severe restraint and allows visual, radiographic, and microscopic examination. The implant test applies a controlled tensile stress to a specimen containing a simulated weld region and is useful for studying hydrogen cracking under defined conditions.

Hot-cracking tests include the Varestraint and Transvarestraint tests, in which augmented strain is applied during solidification. These tests are valuable for comparing alloys and welding conditions, but their severity may exceed that of production joints. The Gleeble thermomechanical simulator reproduces controlled weld thermal cycles and mechanical conditions, allowing separate examination of HAZ transformation, toughness, hardness, and cracking.

Procedure qualification tests commonly include tensile tests, bend tests, Charpy impact tests, hardness surveys, macro- and micro-examination, and specified nondestructive examinations. A passing result demonstrates compliance with the tested range; it does not guarantee immunity under every geometry, restraint, or service condition. Test selection must reflect the governing code, material, thickness, service temperature, loading mode, and failure consequences.

Industrial and Professional Application

In procedure development, the engineer normally begins with material composition, thickness, joint restraint, service requirements, and applicable code. A preliminary heat-input and preheat window is then selected using CE, hardenability information, prior experience, and recognized standards. Consumables are chosen for strength, toughness, corrosion performance, hydrogen control, and dilution behavior. Qualification testing verifies the assumptions.

During production, process control is as important as initial design. Electrode baking and holding, surface cleanliness, preheat measurement, interpass-temperature monitoring, heat-input recording, welder qualification, and traceability provide evidence that the qualified procedure is being followed. Metallurgical failure investigations should preserve samples, map crack locations, examine fracture surfaces, measure hardness, and reconstruct the thermal and restraint conditions.

Digital welding systems increasingly record current, voltage, travel speed, wire feed rate, interpass temperature, and shielding-gas conditions. These data support statistical process control and identify drift before defects become widespread. Thermal simulation and finite-element models can predict temperature fields, phase fractions, residual stress, and distortion, but model outputs require validation through thermocouples, hardness maps, or metallographic observations. Artificial intelligence may assist defect classification and parameter optimization, but it cannot replace sound material knowledge, qualified procedures, or engineering accountability.

Illustrative Case Study: Cracking in a High-Strength Steel T-Joint

Background

A fabricator produced restrained T-joints in a quenched-and-tempered low-alloy steel for a heavy structural assembly. The welds passed visual inspection and initial nondestructive examination, but transverse cracks were detected approximately 24 hours later.

Situation and Analysis

The plate had a relatively high carbon equivalent, and the joint used a thick fillet weld with substantial restraint. Investigation found inconsistent preheat and electrodes exposed to humid air. Hardness measurements showed a high-hardness region in the coarse-grained HAZ. Fractographic examination and crack morphology were consistent with hydrogen-assisted cracking rather than solidification cracking. The delayed appearance, HAZ location, high hardness, and presence of hydrogen supported this diagnosis.

Approach and Solution

The welding engineer introduced a controlled low-hydrogen consumable system, including oven storage and heated holding containers. Joint surfaces were cleaned, a minimum preheat was established from material and thickness requirements, and interpass temperature was monitored. The weld sequence was modified to reduce restraint concentration, and a controlled post-weld holding period was specified before final inspection. A procedure qualification included hardness mapping, bend testing, impact testing where required, and delayed cracking examination.

Outcome

The revised procedure produced lower HAZ hardness and no delayed cracking in qualification and production verification welds. The principal improvement did not arise from one parameter alone; it resulted from controlling hydrogen, cooling rate, restraint, and inspection timing as an integrated system.

Learning Points

The case demonstrates that CE is a screening indicator rather than a complete cracking prediction. It also shows why consumable handling, thermal control, joint design, and delayed inspection must be treated as parts of one weldability strategy.

Challenges, Limitations, and Professional Considerations

Weldability predictions are uncertain because real joints rarely reproduce laboratory conditions. Local chemistry variations, plate thickness, variable restraint, fluctuating heat input, and operator technique can alter results. A low CE does not eliminate cracking, and a high CE does not guarantee failure. Similarly, a small HAZ may not be desirable if it produces excessive hardness.

Common mistakes include selecting filler metal solely by nominal tensile strength, ignoring dilution in dissimilar joints, using heat input without considering t8/5, assuming preheat compensates for wet consumables, and applying a test result outside its validated material or thickness range. Another error is treating inspection as a substitute for prevention. Nondestructive testing detects many flaws, but it cannot restore lost toughness or reliably reveal every delayed metallurgical change.

Professional decisions should comply with applicable welding and pressure-equipment codes, material standards, occupational safety requirements, and environmental controls. Traceability is essential for critical components. Engineers must document assumptions, qualification ranges, deviations, and acceptance criteria. Where failure could threaten life, containment, or the environment, conservative engineering and independent review are justified.

Current Developments and Best Practices

Contemporary weldability engineering increasingly combines physical metallurgy with computational thermodynamics, thermal simulation, digital monitoring, and advanced characterization. Computational tools can estimate phase stability, precipitation, solidification paths, and dilution effects. High-resolution microscopy and automated hardness mapping improve understanding of local damage. Additive manufacturing and hybrid welding introduce new solidification cycles and repeated reheating, making segregation, residual stress, and anisotropy especially important.

Best practice is to:

• characterize the actual material heat, not only nominal grade composition;
• assess carbon equivalent together with thickness, restraint, hydrogen, and cooling time;
• control consumable storage and joint cleanliness;
• define preheat and interpass limits using validated procedures;
• manage dilution in dissimilar and corrosion-resistant overlays;
• qualify procedures using tests relevant to service failure modes;
• monitor production parameters and maintain traceability; and
• investigate defects through integrated metallurgical, mechanical, and process evidence.

Chapter Summary

Welding metallurgy explains how a localized thermal cycle creates distinct weld-metal and HAZ microstructures. Solidification morphology is governed by heat flow, nucleation, growth, alloy constitution, and segregation. Peak temperature and cooling rate control phase transformations, grain growth, hardness, toughness, and softening. Dilution changes deposited chemistry and is critical in dissimilar-metal and overlay welding.

Carbon equivalent provides a useful first estimate of hardenability and hydrogen-cracking risk, but cracking also depends on hydrogen, restraint, thickness, thermal cycle, and stress. Hot cracking develops during solidification, whereas hydrogen-assisted cracking generally develops after cooling in susceptible, stressed regions. Lamellar tearing and reheat cracking require different material and design controls. Weldability tests such as Tekken, implant, Varestraint, Gleeble, and procedure qualification tests provide evidence, but each has defined limits. Reliable practice integrates metallurgy, process control, inspection, standards, and professional judgement.

Key Terms

Carbon equivalent: A compositional index expressing the combined hardening effect of alloying elements relative to carbon.

Coarse-grained HAZ: HAZ region heated far above the austenitizing range, producing significant grain growth.

Dilution: The proportion of deposited weld metal derived from melted base metal.

Equiaxed grains: Grains with approximately equal dimensions in different directions.

Fusion boundary: Interface between solidified weld metal and unmelted base metal.

Heat-affected zone: Base metal whose microstructure or properties change because of welding heat without melting.

Heat input: Thermal energy supplied per unit weld length, adjusted for process efficiency.

Hydrogen-assisted cracking: Delayed cracking caused by diffusible hydrogen, susceptible microstructure, and tensile stress.

Intercritical HAZ: Region heated between lower and upper critical transformation temperatures.

Lamellar tearing: Through-thickness cracking caused by strains acting on inclusion-bearing plate material.

Liquation cracking: High-temperature cracking caused by local melting of grain-boundary constituents.

Macrosegregation: Chemical segregation over relatively large distances in a weld or casting.

Microsegregation: Small-scale compositional variation between dendrite regions.

Preheat: Heating of the base metal before welding to control cooling and hydrogen cracking.

Solidification cracking: Hot cracking occurring during the final stages of weld-metal solidification.

t8/5: Time required for a weld region to cool from 800°C to 500°C.

Thermal cycle: The temperature-time history experienced by a material point during welding.

Weldability: Ability to produce a joint that meets specified fabrication and service requirements.

Review Questions

Short Answer Questions

1. What are the principal regions of a fusion-welded joint?

2. Define dilution and identify two factors that influence it.

3. What information does carbon equivalent provide?

4. State the three essential conditions for hydrogen-assisted cracking.

5. What is meant by t8/5?

Long Answer Questions

1. Explain weld-metal solidification using the concepts of nucleation, epitaxial growth, temperature gradient, and growth rate.

2. Describe the major HAZ subregions in a low-alloy steel and relate each to its thermal cycle.

3. Discuss the mechanisms and prevention of hot cracking and hydrogen-assisted cracking.

4. Explain how heat input, cooling rate, and grain growth influence weld-joint toughness.

5. Compare the purposes and limitations of Tekken, implant, Varestraint, Gleeble, and procedure qualification tests.

Analytical / Application Questions

1. A thick, highly restrained low-alloy steel joint has high CE and inconsistent electrode storage. Predict the dominant cracking risk and propose an integrated prevention plan.

2. A corrosion-resistant overlay shows reduced alloy content near the fusion boundary. Analyse the likely role of dilution and recommend corrective actions.

3. A welded precipitation-hardened aluminium component has a softened HAZ but no visible cracking. Explain the metallurgical cause and propose methods for restoring or accommodating the lost strength.

22. Multiple Choice Questions with Answers

Question 1

Which weld region is most directly associated with epitaxial growth from the base-metal grains?

A. The unaffected base metal

B. The weld-metal region adjacent to the fusion boundary

C. The subcritical HAZ only

D. The far-field residual-stress zone

Correct Answer: B. The weld-metal region adjacent to the fusion boundary

Question 2

A high carbon equivalent in a low-alloy steel most directly indicates increased susceptibility to which condition?

A. HAZ hardening and hydrogen-assisted cracking

B. Porosity caused exclusively by shielding-gas loss

C. Improved ductility during cooling

D. Reduced dilution in the weld pool

Correct Answer: A. HAZ hardening and hydrogen-assisted cracking

Question 3

Which combination represents the essential conditions for hydrogen-assisted cracking?

A. High heat input, low restraint, and coarse inclusions

B. Diffusible hydrogen, susceptible microstructure, and tensile stress

C. Low sulfur, high toughness, and slow solidification

D. Equiaxed grains, low carbon, and no residual stress

Correct Answer: B. Diffusible hydrogen, susceptible microstructure, and tensile stress

Question 4

A weld develops a centreline crack during solidification. Which response is most metallurgically appropriate?

A. Increase impurity levels to widen the freezing range

B. Optimize bead geometry and use filler chemistry resistant to hot cracking

C. Eliminate all preheating and increase restraint

D. Use only delayed post-weld inspection

Correct Answer: B. Optimize bead geometry and use filler chemistry resistant to hot cracking

Question 5

What is the principal effect of excessive dilution in a corrosion-resistant overlay?

A. It can reduce the intended alloy content and alter corrosion performance

B. It always increases corrosion resistance

C. It eliminates fusion-boundary formation

D. It prevents all phase transformations

Correct Answer: A. It can reduce the intended alloy content and alter corrosion performance

Question 6

Which HAZ region is generally associated with the greatest grain coarsening in a low-alloy steel?

A. Subcritical HAZ

B. Intercritical HAZ

C. Coarse-grained HAZ near the fusion boundary

D. Unaffected base metal

Correct Answer: C. Coarse-grained HAZ near the fusion boundary

Question 7

The Tekken or Y-groove test is primarily used to assess susceptibility to:

A. Hydrogen-assisted cracking under restraint

B. Overlay dilution only

C. Austenitic grain refinement during casting

D. Shielding-gas efficiency without mechanical restraint

Correct Answer: A. Hydrogen-assisted cracking under restraint

Question 8

Why does preheating reduce hydrogen-cracking risk?

A. It increases the amount of trapped hydrogen

B. It slows cooling and assists hydrogen diffusion while reducing hard microstructure formation

C. It guarantees zero residual stress

D. It increases weld-metal sulfur content

Correct Answer: B. It slows cooling and assists hydrogen diffusion while reducing hard microstructure formation

Question 9

A T-joint in thick plate fractures parallel to the plate surface near inclusions. Which failure is most likely?

A. Lamellar tearing

B. Centreline solidification cracking

C. Sensitization only

D. Crater porosity

Correct Answer: A. Lamellar tearing

Question 10

Which statement best describes the limitation of carbon equivalent?

A. It completely predicts cracking independently of welding conditions

B. It is useful only for aluminium alloys

C. It is a screening indicator that must be interpreted with hydrogen, restraint, thickness, and cooling rate

D. It directly measures weld-metal dilution

Correct Answer: C. It is a screening indicator that must be interpreted with hydrogen, restraint, thickness, and cooling rate