Welding provides continuous connections by melting the base metal and adding a filler metal to form a single, homogenous structural element. It allows for rigid joints, complex geometries, and highly efficient load transfer without the reduction in cross-sectional net area typical of bolted connections. However, welding requires skilled labor, stringent quality control, and is highly sensitive to environmental conditions (wind, temperature).

• Shielded Metal Arc Welding (SMAW): Often called "stick welding," this process uses a consumable electrode coated in flux. The flux vaporizes to shield the arc. It is highly versatile, portable, and excellent for field work, but it is relatively slow and leaves a slag coating that must be manually chipped off.
• Gas Metal Arc Welding (GMAW): Also known as MIG (Metal Inert Gas) welding. It uses a continuous solid wire feed and an external shielding gas (such as argon or CO2) instead of flux. It is fast, clean, and leaves no slag, but the shielding gas is sensitive to wind, making it best suited for controlled shop fabrication environments.
• Flux-Cored Arc Welding (FCAW): A hybrid process using a continuous tubular wire filled with flux. It can be self-shielded or use an auxiliary shielding gas. It offers extremely fast deposition rates and good wind resistance (if self-shielded), making it the standard process for heavy structural shop and field welding today.
• Submerged Arc Welding (SAW): The arc is completely buried under a mound of granular flux, preventing sparks and spatter. It offers extremely high deposition rates and deep joint penetration but is limited to flat or horizontal welding positions. It is ideal for long, continuous shop welds like plate girder fabrication.

Structural steel can be joined in several distinct ways depending on the geometry of the connection:

• Butt Joint: Two plates are aligned in the same plane and joined edge-to-edge. Typically uses groove welds.
• Lap Joint: Two plates overlap and are joined at their edges. Always uses fillet welds or plug/slot welds.
• Tee Joint: One plate is placed perpendicular to another to form a "T" shape. Uses either fillet or groove welds.
• Corner Joint: Two plates meet at their edges to form a right angle.
• Edge Joint: The edges of two parallel plates are joined together.

• The most common structural weld, accounting for ~80% of all structural connections.
• Deposited in the interior corner formed by two plates meeting at an angle (typically $90^\circ$).
• Assumed to have a right-triangular cross-section.
• Size ($w$): The length of the leg of the largest inscribed right triangle.
• Effective Throat ($t_e$): The shortest distance from the root to the face of the theoretical triangle ($t_e = w \cos 45^\circ = 0.707w$). The strength of a fillet weld is entirely based on shear stress across this minimum throat area, regardless of the direction of the applied load.

• Used primarily for butt joints where plates are aligned, or tee/corner joints requiring high strength. The edges of the plates are often prepared (beveled or V-grooved) to allow the weld to penetrate deeply.
• Complete Joint Penetration (CJP): The weld metal fully penetrates the entire thickness of the base metal. If a "matching" filler metal is used (e.g., E70 for A36/A992 steel), the CJP weld is considered equal to or stronger than the base metal. No complex calculations are needed; you simply design the base metal member.
• Partial Joint Penetration (PJP): The weld only penetrates partway through the thickness. Designed similarly to fillet welds based on an effective throat area ($t_e < t_{plate}$), which depends on the groove angle and welding process.

• Used to connect overlapping plates (lap joints) where there is insufficient edge length for a fillet weld, or to prevent buckling of thin outer plates.
• A round hole (plug) or elongated hole (slot) is punched in the top plate and filled with weld metal to fuse with the bottom plate.
• Strength is calculated based on the shear capacity of the nominal area of the hole ($A_{w}$).

Electrodes are classified by their minimum ultimate tensile strength.

• E70XX: $F_{EXX} = 70$ ksi minimum tensile strength. (The standard "matching" electrode for A36 plates and A992 W-shapes).
• E60XX: $F_{EXX} = 60$ ksi. (Rarely used in modern structural work).
• E80XX, E90XX: High-strength electrodes for specialized high-strength low-alloy steels.

The design strength of a fillet weld is always based on the shear strength of its minimum effective throat area, even if the load is applied in tension or compression relative to the weld axis.

• $\phi = 0.75$ (LRFD).
• The $0.60$ factor converts tensile strength to shear strength (von Mises criterion).

Unlike bolted connections, which can often be visually inspected for proper installation (e.g., checking for the correct number of bolts and using a torque wrench), internal weld quality is impossible to verify by eye alone.

• Porosity: Small gas pockets or voids trapped inside the weld metal, usually caused by poor shielding gas coverage or moisture on the electrodes.
• Slag Inclusions: Solid non-metallic material (flux) trapped within the weld metal or between the weld and base metal. Common in multipass SMAW or FCAW welding if the slag is not properly chipped away between passes.
• Undercut: A groove melted into the base metal adjacent to the toe of the weld and left unfilled by weld metal. This creates a severe stress concentration.
• Lack of Fusion / Incomplete Penetration: The weld metal fails to completely fuse with the base metal or fails to reach the required depth in a groove weld. This drastically reduces the strength of the joint.

To detect these defects, various NDT methods are employed:

• Visual Inspection (VT): The first line of defense. Checks for proper size, profile, and surface defects like severe undercut or surface porosity.
• Dye Penetrant Testing (PT) & Magnetic Particle Testing (MT): Used to detect microscopic surface-breaking cracks or defects.
• Ultrasonic Testing (UT): Uses high-frequency sound waves to detect internal flaws (lack of fusion, deep slag inclusions). It is the primary method for inspecting Complete Joint Penetration (CJP) groove welds.
• Radiographic Testing (RT): X-raying the weld to produce an image of internal defects. Highly accurate but expensive and hazardous.

Fillet welds loaded transversely (perpendicular to their longitudinal axis) are significantly stronger (approximately $50\%$ stronger) than welds loaded longitudinally (parallel to their axis). For a linear weld group loaded at an angle $\theta$ relative to its longitudinal axis, AISC specifies a directional strength increase factor.

However, transverse welds ($\theta = 90^\circ$) are less ductile and experience rupture at smaller deformations compared to longitudinal welds.

A welded connection is only as strong as its weakest link. While the deposited weld metal is typically stronger than the base metal being joined, the strength of the connected plates adjacent to the weld must be checked independently. The connection capacity is the lesser of the weld metal shear strength and the base metal shear strength.

When connecting unsymmetrical members (like single angles connected by one leg) subjected to axial forces, the center of gravity of the member does not naturally align with the geometric center of a symmetrical weld group. If equal-length welds were placed on both edges, the applied load would create an eccentric moment, inducing secondary bending stresses that could cause premature joint failure.

To "balance" the weld and eliminate eccentricity, the designer proportions the lengths of the longitudinal fillet welds on the heel and toe of the angle so that the centroid of the resistance from the weld group aligns perfectly with the centroid of the applied force.

This is determined by applying static equilibrium ($\sum M = 0$) about each weld line. The weld farthest from the center of gravity (the heel) will be longer, and the weld closest to the center of gravity (the toe) will be shorter.

When an external load is applied eccentrically to a weld group (e.g., a bracket welded to a column flange supporting an offset load), the weld must resist both direct shear ($P/L$) and a twisting moment ($Pe$).

The preferred, exact method in the AISC Manual for analyzing eccentric weld groups. It is highly non-linear. It assumes the weld group rotates rigidly around an "instantaneous center." The deformation of each weld segment is proportional to its distance from the ICR.

Because the equations are extremely complex to solve by hand, designers rely entirely on the tabulated coefficients ($C$) in AISC Manual Part 8. The capacity is simply $R_n = C C_1 D l$, where $C$ is the coefficient from the table based on geometry, $C_1$ adjusts for electrode strength, $D$ is the weld size in sixteenths, and $l$ is the length.

A simplified, conservative, traditional approach. It assumes the load translates to the centroid of the weld group, causing uniform direct shear ($f_v = P/A_{we}$), plus a twisting moment that causes torsional shear stresses proportional to the distance from the centroid ($f_t = M r / I_p$). These vectors are added geometrically at the most critical point to find the maximum stress.

Depends solely on the thickness of the thicker part joined. A small weld on a massive, thick plate will cool so rapidly (due to the plate acting as a heat sink) that the weld metal becomes brittle and cracks, or fails to fuse properly.

• $t_{thicker} \le 1/4 \to w_{min} = 1/8$ in.
• $1/4 \lt t_{thicker} \le 1/2 \to w_{min} = 3/16$ in.
• $1/2 \lt t_{thicker} \le 3/4 \to w_{min} = 1/4$ in.

Depends on the thickness of the thinner part, specifically when welding along the edge of a plate. If the weld is too large, the heat will melt away the edge of the plate, reducing its effective thickness.

• Along edge of material $\lt 1/4$ in thick: $w_{max} = t$.
• Along edge of material $\ge 1/4$ in thick: $w_{max} = t - 1/16$ in. (The 1/16" gap ensures the inspector can verify the corner wasn't melted off).