3.3 Product Forms: Welds, Castings, Forgings, Wrought, In-Service Damage

Product form tells you which family of discontinuities is even possible. The radiographic general outline explicitly includes manufacturing processes and associated discontinuities; ultrasonic outlines evaluate base-material product forms and weldments; magnetic particle and visual outlines list discontinuity types; the specific exam's Applications items mix all of them. If you look for a forging lap in a sand casting, or a hot tear in rolled plate, you are already on the wrong map.

Inherent, processing, and service

Sort every discontinuity into one of three origin classes before you name it.

A stringer in bar is an inherent inclusion that became a processing-shaped discontinuity. A lamination in plate is usually pipe or inclusions flattened by rolling. The exam will accept either inherent origin with processing morphology or the product-form name. What it will not accept is calling a lamination a weld defect.

Castings

Castings solidify in a mold. The last metal to freeze, the gates and risers, and any turbulence in the pour write the discontinuity list.

• Shrinkage is a cavity left when liquid metal contracts and is not fed by a riser. It is irregular or dendritic, often in the thermal center. Radiographic and ultrasonic testing are the volumetric methods; shrinkage is not a penetrant or magnetic-particle condition unless it breaks the surface.
• Gas porosity is trapped mold gas or dissolved gas coming out of solution. Voids are spherical or elongated. Radiography is the classic map; surface pores are visual or penetrant calls (and magnetic particle calls if the casting is ferromagnetic).
• Inclusions are sand, slag, or oxides folded into the metal. They can be irregular and may cluster near the cope or in gates. Radiography and ultrasound find many of them.
• Hot tears form while the casting is still hot and weak, usually at section changes or hot spots where contraction is restrained. They are ragged cracks. Surface-breaking tears: visual, penetrant, and magnetic particle testing. Internal tears: radiography and ultrasound.
• Cold shuts form when two streams of metal meet but do not fuse, often with an oxidized interface. Many cold shuts reach the surface and are excellent visual, penetrant, and magnetic-particle targets. A completely internal cold shut behaves like a planar bond-line and is an ultrasonic or radiographic problem.

Castings also carry chaplets, core shift, and misrun — visual and dimensional checks catch many of those before a volumetric method is applied. A specific-exam stem that says steel sand casting plus irregular dendritic cavity in the last region to freeze is asking for shrinkage, not a forging burst.

Forgings and wrought products

Forging, rolling, and extrusion work metal in a direction. Discontinuities elongate in the working direction and often lie parallel to the surface in plate.

• Laps are folds of metal squeezed into the surface, usually oxidized, so they will not weld shut. They are surface or near-surface. Visual, penetrant, and magnetic particle testing (ferromagnetic) are primary. Ultrasound may see a larger lap as a near-surface reflector.
• Seams are elongated surface discontinuities, often from ingot cracks or trapped scale that rolled out along a bar. They run with the length. Magnetic particle testing on bar stock is the textbook method; penetrant testing works on nonmagnetic alloys.
• Bursts are ruptures from forging that is too severe, too cold, or performed on a dirty ingot. They may be internal (forging burst) or surface (forging crack). Ultrasound is the internal screen; magnetic particle or penetrant testing catch surface breaks.
• Laminations are separations in plate or pipe wall, parallel to the rolled surfaces, from pipe or inclusions flattened during rolling. Straight-beam ultrasonic testing is the method of choice. Radiography is a poor primary method because the plane is usually parallel to the film and presents almost no thickness change. Penetrant or magnetic particle testing see a lamination only at a cut edge or if it breaks out.
• Stringers are elongated nonmetallic inclusions aligned with the working direction. They may leak flux if they break a ferromagnetic surface; ultrasound finds many internal stringers as short reflectors along the grain.

Wrought product form also includes plate, bar, pipe, and fittings. The Level II should expect laminations in plate, seams in bar, and, in seamless pipe, residual centerline conditions from the piercing mill — all straight-beam ultrasonic stories unless they break the surface.

Welds

Welds add a second solidification and a heat-affected zone (HAZ). Learn the names the codes use.

• Lack of fusion (LOF) is a planar failure of weld metal to fuse to base metal or to a previous pass. It sits on a bevel or interpass boundary. Angle-beam ultrasonic testing is the strongest primary method. Radiography finds lack of fusion only when the plane is open and aligned with the beam. Visual, penetrant, or magnetic particle testing see it only if it breaks an accessible surface.
• Lack of penetration (LOP) or incomplete penetration is failure to fill or fuse through the required throat or root. A visible root can be a visual call. Radiography with the beam along the root land, and ultrasound, are the volumetric options.
• Slag inclusions are trapped flux. They are irregular or linear along the weld. Radiography images them well. Ultrasound reflects from them. Surface-connected slag is a visual call.
• Porosity is trapped gas in weld metal — scattered, clustered, or piping. Radiography is the reference map. Surface pores: visual or penetrant testing.
• Undercut is a groove melted into the base metal at the toe or root. Visual testing is primary. Penetrant or magnetic particle testing confirm. Do not start by ultrasonically examining an undercut you can see.
• Hot cracks (solidification cracks) form as the weld metal freezes — centerline cracks, crater cracks, and some fusion-line tears. They may be visual, penetrant, or magnetic-particle calls if open; radiographic or ultrasonic calls if internal.
• Cold cracks, including hydrogen-assisted and many HAZ cracks, form after the weld has cooled, sometimes hours later. Toe cracks and underbead cracks are the usual pictures. Magnetic particle testing on carbon steel and penetrant testing on any metal are the surface tools; ultrasound sizes depth. Delayed cracking is why some procedures impose a wait time before final surface examination.
• Crater cracks are star-shaped cracks in an improperly filled crater. They are surface and are visual, penetrant, or magnetic-particle calls.

AWS D1.1 and the construction codes that point to ASME Section V treat these names as evaluation categories, not as slang. Do not call a slag line a seam, and do not call a heat-affected-zone toe crack a cold shut.

In-service damage

Once the product is in use, a new family appears. The specific exam (General Industry or Pressure Equipment) expects awareness, not a metallurgy thesis.

• Fatigue is crack growth under cyclic load. It almost always starts at a surface stress raiser: a weld toe, a thread root, a corrosion pit, or a sharp fillet. Visual testing plus magnetic particle or penetrant testing is the screening pair. Ultrasound follows for depth. Early fatigue is a terrible radiographic target.
• Corrosion removes metal or pits the surface. General wall loss is an ultrasonic thickness (or radiographic profile) problem. Pitting may be visual and, if you need to see whether pits are crack-like, penetrant testing.
• Erosion and flow-accelerated attack cut metal downstream of fittings, in elbows, and at control valves. Ultrasonic thickness grids find it; visual testing finds it when the inside diameter is accessible.
• Hydrogen damage includes blistering, hydrogen-induced cracking (HIC) in stepwise mid-wall planes, and hydrogen-embrittlement cracks at hard spots. Ultrasound (including compression-wave mapping) is the internal tool; magnetic particle or penetrant testing catch surface breaks.
• Creep is time-dependent strain at high temperature. Early creep may show no conventional visual, penetrant, magnetic particle, radiographic, or ultrasonic response. Advanced creep can produce microcracking and cavities that surface methods and replicas begin to see. Treat creep as an awareness item: do not claim a clean yoke examination proves a high-temperature header has remaining life.

Method-to-family map

Realistic specific-exam stems

A steel sand casting is radiographed and shows an irregular dendritic cavity in the last region to freeze, plus scattered round voids under the cope. That pair is shrinkage plus gas porosity — inherent, volumetric, radiographic territory — not a forging burst.

Rolled plate is straight-beam tested from the surface and returns a mid-wall echo with a reduced back reflection. That is a lamination, not lack of fusion (there is no weld) and not a cold shut (that is a casting term).

A carbon-steel girth weld in a pressure vessel, examined after a hydrotest wait time, shows a sharp linear magnetic-particle indication in the heat-affected zone at the toe. That is the picture of a delayed cold / hydrogen crack, not slag (slag is volumetric on a radiograph) and not a seam (seams live in wrought base metal along the working direction).

An elbow downstream of a control valve has lost 30 percent of nominal wall on an ultrasonic thickness grid and shows no crack-like magnetic-particle indication. That is erosion or flow-accelerated metal loss, not fatigue, and NDT still has not computed remaining life — it has measured remaining thickness.

Keep the product-form dictionary attached to the method map. The Level II who can say this is a lamination, so reach for straight-beam ultrasound, not penetrant testing, is the candidate the specific exam is written to pass.