ME 538: Welding Design, Fabrication and Quality Control
Estimated study time: 10 minutes
Table of contents
Sources and References
- Kou, Welding Metallurgy, 2nd ed., Wiley.
- Messler, Principles of Welding: Processes, Physics, Chemistry, and Metallurgy, Wiley.
- Masubuchi, Analysis of Welded Structures, Pergamon.
- AWS Welding Handbook, 10th ed., American Welding Society.
- ASME Boiler and Pressure Vessel Code, Section IX.
- Hellier, Handbook of Nondestructive Evaluation, 2nd ed., McGraw-Hill.
Chapter 1: Welding Processes and Metallurgical Consequences
Welding joins metals by melting (fusion welding) or by plastic flow under pressure (solid-state welding). The process chosen determines heat input, cooling rate, consumable chemistry, distortion, and residual stress — all of which affect the mechanical performance of the joint.
1.1 Survey of Processes
Shielded metal arc welding (SMAW), gas metal arc welding (GMAW), gas tungsten arc welding (GTAW), flux-cored arc welding (FCAW), and submerged arc welding (SAW) cover the bulk of structural fabrication. Electron-beam and laser welding deliver very high power density and low heat input. Solid-state processes such as friction-stir welding avoid melting and its attendant solidification cracking.
The key process variable for metallurgy is specific heat input,
\[ H = \eta \frac{V I}{v}, \]where \( V \) is voltage, \( I \) is current, \( v \) the travel speed, and \( \eta \) the arc efficiency. Heat input governs peak temperature, fusion-zone size, and, via the cooling rate, the phase transformations in the heat-affected zone (HAZ).
1.2 The Weld Thermal Cycle
Rosenthal’s moving-point-source solution for a thick plate is
\[ T - T_0 = \frac{\eta V I}{2 \pi k r} \exp\!\left[-\frac{v(r + x)}{2 \alpha}\right], \]with \( k \) the thermal conductivity and \( \alpha \) the diffusivity. At a given point away from the weld axis, the temperature rises rapidly, reaches a peak, and decays toward ambient. The cooling rate between \( 800 \) °C and \( 500 \) °C (the \( t_{8/5} \) time) is the customary metric for steel HAZ behaviour; faster \( t_{8/5} \) promotes martensite formation.
Chapter 2: Design of Welded Joints
2.1 Joint Types and Preparations
Butt, lap, T-, corner, and edge joints are realized with square, single-V, double-V, single-bevel, or J-groove preparations. The preparation reflects thickness, access, and cost: complete-joint-penetration welds carry the full plate strength; partial-penetration welds accept a defined defect depth in exchange for lower cost and distortion; fillet welds handle corner and lap geometries.
2.2 Fillet-Weld Strength
For a fillet weld of leg length \( w \), the throat is \( t = w/\sqrt{2} \). Static strength is
\[ F = t L \tau_{allow}, \]with allowable shear \( \tau_{allow} \) scaled from electrode tensile strength and reduced for the loading direction per code (AISC, CSA, Eurocode). Weld groups under eccentric shear are solved by the instantaneous-centre-of-rotation method or by linear elastic superposition, depending on allowed design philosophy.
2.3 Fatigue of Welded Joints
Welds are fatigue-critical because toe stress concentrations and residual tension conspire against the base metal’s endurance. Design codes (AASHTO, IIW, ASME) classify joint geometries into S–N categories, each with an allowable stress range versus cycles curve. Basic formula,
\[ \Delta \sigma = C N^{-1/m}, \]typically has \( m \approx 3 \). Weld toe grinding, hammer peening, or high-frequency mechanical impact improves category by one or two levels by reshaping toes and introducing compressive residual stress.
2.4 Design for Alloy Class
Plain-carbon and low-alloy steels are weldable over a wide range with appropriate preheat and consumables. The carbon equivalent
\[ CE = C + \frac{Mn}{6} + \frac{Cr+Mo+V}{5} + \frac{Ni+Cu}{15} \]correlates with hardenability and, above about 0.45, signals the need for substantial preheat and low-hydrogen processes. Aluminium alloys require careful choice of filler (4xxx or 5xxx) to avoid cracking and to suit post-weld heat treatment or natural aging. Austenitic stainless steels demand control of ferrite content and heat input to avoid sensitization.
Chapter 3: Residual Stresses and Distortion
3.1 Origin of Residual Stress
The weld metal and HAZ undergo thermal expansion and contraction while restrained by surrounding cold material. On cooling, the once-molten zone cannot freely contract; it locks in tensile residual stress that is typically near the yield stress of the base metal in the welding direction. Transverse and through-thickness stresses depend on joint geometry and restraint.
3.2 Distortion Modes
Distortion is classified as transverse shrinkage, longitudinal shrinkage, angular distortion, bowing, buckling, and twisting. Angular distortion in a single-V butt weld arises from the unequal volume of weld metal above and below the neutral axis; it can be mitigated by backstep welding, balanced passes, or prebending the plates.
3.3 Control Strategies
Control combines process choice (low heat input, multi-pass sequencing), geometry (symmetric joints, balanced filler volumes), fixturing (robust clamping, strongbacks), and post-weld treatment (stress relief, vibratory stress relief, peening). Predictive finite-element models using thermo-elastic-plastic analysis with transformation-induced plasticity allow the designer to evaluate alternatives before metal is cut.
\[ \sigma_{res,max} \approx \sigma_y, \qquad \delta_{angular} \propto \frac{Q}{t^2}, \]approximate scaling relations that serve for quick comparative assessments.
Chapter 4: Quality and Quality Control
4.1 Welding Procedure and Performance Qualification
Codes require a written welding procedure specification (WPS) supported by a procedure qualification record (PQR) in which a test coupon was welded under the proposed parameters and subjected to mechanical testing. Welders themselves are qualified by welding a coupon per a separate performance-qualification standard. ASME Section IX is the standard reference.
Essential variables for the WPS include base-metal group, filler-metal class, preheat range, interpass temperature, heat input, and position. Changing an essential variable requires requalification; changing a non-essential variable does not.
4.2 Acceptance Criteria
Codes such as AWS D1.1 (structural steel), ASME B31.3 (process piping), and ASME Section VIII (pressure vessels) specify defect types and sizes. Typical disallowed defects include cracks (never allowed), lack of fusion, lack of penetration (except in partial-penetration welds), and porosity above a defined cluster density. Slag inclusions and undercut are size-limited.
4.3 Nondestructive Examination
Nondestructive examination (NDE) methods include visual, radiographic, ultrasonic, magnetic-particle, dye-penetrant, and eddy-current techniques. Each excels on a particular defect class and geometry.
Visual testing (VT) is the baseline and catches most gross surface defects. Dye penetrant (PT) reveals surface-breaking cracks on any material; magnetic particle (MT) is more sensitive on ferromagnetic materials and can find near-surface defects. Radiographic testing (RT) produces a shadow image of the weld interior; it is ideal for volumetric defects (porosity, slag) but less sensitive to planar cracks aligned with the beam. Ultrasonic testing (UT), particularly phased-array and time-of-flight diffraction (TOFD), excels at planar defects and gives good through-thickness information. Eddy-current testing (ET) is surface-oriented and useful for in-service inspection and tubing.
Chapter 5: Weldability and Failure Avoidance
5.1 Hydrogen-Assisted Cracking
In high-strength ferrous welds, atomic hydrogen from moisture, oils, or cellulosic electrodes diffuses into the HAZ. Combined with a susceptible microstructure (high-hardness martensite) and tensile residual stress, it produces cold cracks hours or days after welding. Prevention uses low-hydrogen consumables baked and stored dry, sufficient preheat to slow cooling and permit diffusion out of the joint, and post-weld hydrogen bake-out for heavy sections.
5.2 Solidification and Liquation Cracking
Hot cracking occurs at the end of solidification when thin liquid films are drawn apart by thermal strain. Susceptibility is high when solidification ranges are wide and low-melting constituents (sulphides, phosphides in steels; silicon eutectics in aluminium) collect at grain boundaries. Remedies include controlled composition, lower heat input, and joint preparations that reduce restraint.
5.3 Post-Weld Heat Treatment
Stress relief at 580–620 °C for carbon steels reduces residual stress by diffusion-controlled creep at temperature and tempers any martensitic HAZ. Solution and aging treatments restore heat-treatable aluminium alloys to a controlled temper. Normalizing or reheat refines coarse HAZ grains in some ferritic steels. Each treatment is specified in the WPS and verified by mechanical testing on the PQR coupon.
