Choosing a Nondestructive Test Method for Steel

Start from the flaw, not from the instrument

Nondestructive evaluation is a family of activities rather than a single test. The published guide divides it into nine areas, of which seven are useful on steel. They are flaw detection, leak detection, dimensional metrology, locating a component inside an assembly, characterizing microstructure, estimating mechanical and physical properties, and measuring stress or strain. The remaining two, signature analysis and chemical composition, are handled by other methods and are outside this volume.

Almost every question that reaches a tool steel supplier falls under flaw detection, and the published advice for picking a method is to fix six things before any instrument is chosen. They are the reason for the test, the type of flaw, the size and orientation of the flaw that will be rejected, where the flaw is expected to sit, the size and shape of the part, and the characteristics of the material.

The reason falls into three cases. In-process inspection checks a part after a fabrication step, final inspection clears it for use, and in-service inspection decides whether a part already in service can continue. Once the reason is set, the rejectable flaw has to be defined, and it is usually defined by a code or a purchase specification rather than by the shop floor. That rejectable flaw then drives everything else.

Volumetric flaws and the methods that find them

A volumetric flaw is one that can be described by three dimensions or by a volume. Porosity, inclusions, slag, shrinkage and corrosion all fall into this class, and they change the density or the cross section of the steel in a way that a through-thickness method can detect. The two columns below are independent lists rather than matched pairs. The first names the volumetric flaws, and the second names every method in the volume that has been used to detect them.

Volumetric flaw classification and the detection methods applied to it, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17. Entry methods marked surface react only to a flaw that reaches the surface, and the rest are stated against the depth they can reach where the source gives it.

Planar flaws need a different set of methods

A planar flaw is thin in one dimension and larger in the other two, so it carries almost no volume to change the density of the steel. A seam, a lamination and a crack are all planar, and the methods that find them are largely different from those that find a void. This is why a single test rarely closes the question, and why the same bar can pass one method and fail another on a different flaw.

Planar flaw classification and the detection methods applied to it, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17. Heat treatment cracks, grinding cracks and fatigue cracks in this list are the ones a tool steel user meets most often.

Surface flaws and interior flaws

The next split is position. A flaw that breaks the surface is accessible to a chemical or a magnetic method, while a flaw that sits inside the steel needs a wave that travels through it. The two columns below again hold independent lists, one for methods that reach a surface flaw and one for methods that reach an interior flaw. A short list appears in both, because it works in both places.

NDE methods for the detection of surface and interior flaws, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17. Magnetic particle is marked as limited use for interior flaws because its useful depth is small, and the entries marked possible are the borderline ones in the source.

How thick the part is, and how complex

Penetrating power sets a ceiling on the part size each method can handle, and the published guide gives approximate limits. The thickness figures are approximate because the exact value depends on the physical properties of the material being examined, and any method that suits a thick object can also be used on a thin one. Neutron radiography is the single exception, since it is not useful on most thin objects.

Comparison of NDE methods by size of the object to be evaluated, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17. A method that suits a thick object can also be used on a thin one, except neutron radiography.

Shape is the second constraint, and it runs in the opposite direction to size. The methods that reach deepest and find the smallest flaws need the simplest shape, and the methods that tolerate a complex shape work only at the surface. The list below runs from the simplest shape at the top to the most complex at the bottom.

Comparison of NDE techniques by shape of the object to be evaluated, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17.

What the material itself allows

Every method rests on a physical property of the steel, and if that property is absent the method cannot be used no matter how good the flaw definition is. Liquid penetrant needs the flaw to break the surface, magnetic particle needs the steel to be magnetic, eddy current needs it to be conductive or magnetic, and radiography needs a difference in thickness, density or composition to cast a shadow. A nonmagnetic stainless grade, or an austenitic one, removes magnetic particle from the list before anything else is considered.

NDE methods and the material characteristic each one depends on, reproduced from the Guide to Nondestructive Evaluation Techniques in ASM Handbook, Volume 17.

Because each method has its own blind spot, the guide recommends combining methods with complementary reach. Ultrasonic inspection and radiography are the standard pair, since ultrasonic responds strongly to a planar crack that radiography may pass and radiography responds strongly to a void that ultrasonic may miss. What comes back from the test is only as good as the definition of the rejectable flaw that was written down first.

Grades and stock

The flaw types in these lists are the ones a tool steel user meets on real bar and die work, and they are described one by one on the steel bar defects page, which covers pipe, seams, laps and chevrons with an illustration of each.

A nondestructive test sits on top of the chemical and metallographic checks rather than replacing them, and the two halves together are what a mill certificate reports. The chemical side is set out on the page on how steel chemistry is tested, and the metallographic side on the inclusion and grain size page.

Where a part has already failed, the surface marks left by a crack are usually more informative than a fresh test of a sound part, and they are catalogued on the fracture surface features chart. The defects that grindability and heat treatment leave behind are collected on the decarburization and stock removal page.

Source: ASM Handbook, Volume 17, Nondestructive Evaluation and Quality Control, ASM International, 1989.