Study ASNT NDT Level III Electromagnetic Testing by treating every physics fact as a decision input: skin depth sets frequency, phase position carries depth information, coil configuration follows from geometry, and material properties determine which electromagnetic technique is even usable. Two worked scenarios and a calculation drill turn this into a four-week preparation sequence.
What the ASNT NDT Level III ET credential covers and how the exams fit together
ASNT NDT Level III certification pairs a basic examination with the ET-specific examination, both computer-based, and is accredited under ISO/IEC 17024. Hands-on practical testing belongs to the separate ASNT 9712 program, not this one.
ASNT delivers Level III certification through two distinct programs. The ASNT NDT Level III route consists of computer-based examinations, including the ET-specific exam. The ASNT 9712 Level III route adds hands-on practical examinations and an assessment of your ability to prepare instructions or procedures. Your circumstances decide which route fits: if your employer's internal certification program covers practicals, the ASNT NDT route matches it; if you need a designation recognized for contractor work across multiple countries, the 9712 structure may fit better.
Because the ASNT NDT Level III ET examination is knowledge-based, preparation should concentrate on the electromagnetic testing body of knowledge itself: electromagnetic principles, technique families, signal interpretation, and the judgment needed to write and review procedures. Administrative matters, such as applications, documented experience, and scheduling, are handled by ASNT and do change, so verify current requirements directly on the ASNT certification pages rather than relying on third-party summaries of them.
Skin depth is the control knob: turning the formula into a frequency decision
The standard depth of penetration, delta equals one over the square root of pi times f times mu times sigma, defines where eddy currents flow. Effective inspection depth is commonly taken as about three delta, so frequency choice defines what you can detect.
Each term carries a decision. Raising frequency shrinks delta, which improves sensitivity and resolution near the surface but abandons deeper material. Lowering frequency extends penetration at the cost of sensitivity and a fatter, harder-to-read signal. Permeability enters under the square root, so a ferromagnetic material has a far smaller delta than a nonmagnetic conductor at the same frequency. Phase also matters: the eddy current phase lags by roughly one radian per standard depth of penetration, which is exactly why the phase position of a signal carries depth information.
Run the numbers once by hand so the magnitudes stay intuitive. For copper, with conductivity near 5.8 x 10^7 S/m, delta is about 0.21 mm at 100 kHz and about 0.66 mm at 10 kHz. A tenfold frequency increase reduces delta by roughly a factor of three, not ten, because of the square root. Now test your decisions against that: if a task requires detecting a flaw 1.5 mm below a copper surface, a 100 kHz setup inspecting to roughly three delta (about 0.63 mm) cannot reach it, and the frequency must drop substantially even though surface sensitivity will suffer as a trade.
Impedance plane variables: separating lift-off, conductivity, and defect signals
The eddy current response is a vector in the impedance plane, not a single number. Lift-off, conductivity, permeability, and defect geometry each move the operating point in characteristic directions, and frequency changes how distinguishable those directions are.
Plot the coil's resistance and inductive reactance as a point. As the probe approaches a conductive part, that point traces a path; for nonmagnetic metals of different conductivity the endpoints fall along a conductivity curve. Lift-off, the coil-to-surface distance, produces a direction of movement that is largely independent of the metal, while a defect signal generally departs from the lift-off direction. That angular separation is the entire basis of phase analysis: an operator or an instrument judges indication depth by where the signal sits in phase, so anything that scrambles the phase map, such as wobble or a curved surface changing lift-off, corrupts depth estimates too.
Two Level III-level consequences follow. First, frequency mixing combines responses at two frequencies so that the lift-off contribution cancels while material and defect signals remain, which is how wobbling probes and support-plate backgrounds are suppressed in tube work. Second, permeability variation in ferromagnetic material produces signal changes that can dominate the conductivity-based information you are trying to measure; this is why ordinary eddy current on carbon steel demands either magnetic saturation or a different electromagnetic technique family, a decision that belongs in the procedure, not in the operator's improvisation.
Coil configurations and technique families: matching the setup to the geometry
Absolute coils respond to the total condition under the coil; differential coils respond to change along the scan. Geometry dictates the arrangement, and material dictates which electromagnetic technique family is usable at all.
Absolute setups read the full state of the material, which makes them natural for conductivity sorting, thickness, and slow property gradients, but they also record drift, temperature effects, and lift-off change as signal. Differential pairs cancel anything that changes slowly along the scan, which is why production tube lines favor them; the price is that a defect changing gradually along the scan direction, such as a long tapering flaw, may produce only a weak differential response. Encircling coils give full circumferential coverage at line speed but favor certain orientations; bobbin coils inspect tubing from the inside quickly but see primarily axially symmetric changes; rotating probes trade speed for the ability to resolve flaw orientation.
Eddy current is only one member of the electromagnetic method family. Remote field testing uses lower-frequency through-wall fields and is applied to ferromagnetic tubing where ordinary eddy current penetration collapses; magnetic flux leakage detects anomalies through the leakage field near ferromagnetic surfaces; AC field measurement infers surface cracks from perturbations of an induced surface field. These differ in physical basis, sensitivity direction, and material constraints, and selecting among them, with documented justification, is precisely the kind of judgment a Level III procedure must capture.
Use the table below as a decision scaffold, then justify each choice in your own words for your own applications.
- Treat the table as a starting map: real procedures must reconcile it with the actual part geometry, access, and material condition.
- For every row, be able to state why the alternative rows fail before you claim the selected row succeeds.
| Geometry and objective | Typical candidate technique | Signal behavior to expect | Main caution |
|---|---|---|---|
| Surface cracks on nonferrous plate or welds | Absolute surface (pancake) probe eddy current | Defect signals depart from the lift-off direction; phase relates to depth | Lift-off variation mimics signal changes; control probe handling or apply mixing |
| Bolt and fastener holes | Rotating probe eddy current | Signal shape repeats around the hole; orientation discrimination by scan | Slower coverage; reference notch orientation must match expected flaw orientation |
| Long production runs of nonferrous tube | Encircling differential coil | Fast full coverage; responds to change along the line | Gradual, long-changing flaws can give weak differential response |
| In-service nonferrous heat exchanger tubing from the ID | Internal bobbin differential eddy current, with frequency mixing | ID and OD indications separate in phase; mixing suppresses support-plate background | Axially oriented cracking is better verified with a rotating probe |
| Ferromagnetic tubing where penetration is inadequate | Remote field testing | Through-wall information carried largely in phase | Different signal interpretation; verify suitability for the wall thickness and flaw type |
| Conductive coating or nonconductive coating thickness | Lift-off based eddy current measurement | Lift-off is used deliberately as the measurand | Requires frequency and calibration matched to the coating and substrate |
Worked scenario one: selecting a technique for in-service nonferrous heat exchanger tubing
For thin-wall nonferrous tubing inspected from the inside, a differential bobbin at a frequency giving roughly full-wall penetration, plus frequency mixing, provides fast coverage; a rotating probe verifies crack-like indications before disposition decisions.
Scenario: stainless tubing, roughly 1.2 mm wall, inspected from the inside, with support plates at intervals and a requirement to grade pit depth on both surfaces. A plausible mistake is to select an absolute bobbin at a low frequency 'for penetration.' The absolute response records probe drift and property variation along the tube as apparent indications, the low frequency compresses the phase spread between ID and OD flaws so depth grading becomes unreliable, and the support plates flood the trace with background signal. The result is a data set that looks busy but cannot actually support depth calls.
The better sequence starts from the physics: compute delta for the alloy, choose a frequency where the wall is well within the effective penetration range, then confirm on reference standards containing ID and OD notches that the two responses separate cleanly in phase, because that separation is what depth grading is built on. Add a second frequency and mix to cancel lift-off and support-plate contributions, and specify a rotating probe pass to characterize any crack-like indications, since a bobbin cannot reliably resolve orientation. Why it matters: the procedure's calibration standards and frequency choice are not paperwork; they are what makes an ID versus OD depth call defensible.
Worked scenario two: reviewing an ET procedure for a ferromagnetic weld inspection
A procedure applying ordinary eddy current to unsaturated carbon steel must address permeability variation. The reviewer's defensible options are magnetic saturation or a technique family suited to ferromagnetic materials, stated explicitly in the procedure.
Scenario: a contractor's procedure proposes a surface-probe eddy current examination of fatigue cracks in carbon steel welds, with a frequency chosen 'for sensitivity.' A plausible mistake in review is to approve it because the eddy current physics and the frequency reasoning look correct in isolation. What the procedure never confronts is that permeability varies across the weld and heat-affected zone, so the impedance plane carries large, spatially varying permeability-driven signals that can sit in the same region as crack responses, burying the crack information the examination exists to find.
The better review asks three questions the procedure must answer: does it specify magnetic saturation of the inspection volume, or does it name a technique family designed for ferromagnetic conditions such as remote field or AC field measurement; does it demonstrate, on reference welds, that crack responses are distinguishable from material noise under the stated settings; and does it pin down frequency, lift-off control, and calibration elements rather than leaving them to operator judgment. Why it matters: procedure approval is a core Level III responsibility, and the failure mode here is invisible in the mathematics, surfacing only when the procedure is tested against real, variable material.
A four-week preparation sequence with a calculation drill and self-check rubric
Spend weeks one and two on electromagnetic fundamentals and calculations, week three on technique families and coil selection, and week four on scenarios and drafting procedure outlines. Treat the readiness checks as milestones, not predictions of any exam outcome.
Weeks one and two: skin depth, phase lag, Lenz's law, and the impedance plane, with hand calculations for several metals and frequencies until magnitudes feel intuitive. Week three: coil types, absolute versus differential behavior, display and mixing concepts, and the boundaries between eddy current, remote field, flux leakage, and AC field measurement. Week four: run the scenarios above with your own part geometries, then draft one-page procedure outlines, frequency, coil, calibration standards, lift-off control, for two different applications without a template.
Practical exercise, the depth-drill: compute delta for copper, aluminum, and steel at 10, 60, and 100 kHz; for each result state the expected phase lag at two delta; sketch an impedance plane and mark the lift-off direction, the conductivity curve, and where a shallow versus a deep defect would sit. Expected observations: delta shrinks by roughly a factor of three per tenfold frequency rise; steel's delta at the same frequency is far smaller than copper's because permeability enters under the square root; the lift-off direction stays consistent while defect phase rotates with depth. If your sketch shows defects moving along the lift-off direction instead of rotating away from it, revisit section three before moving on.
Readiness checks: you can compute and sanity-check delta without notes; you can state the phase lag per standard depth and use it to explain ID versus OD discrimination; you can sketch the impedance plane with all three variable directions; you can justify a coil and frequency for a given geometry in under a paragraph; you can name the conditions under which ordinary eddy current is unsuitable for ferromagnetic material and state the two remedies; you can outline a procedure's calibration elements unprompted. If you fall short on one check, return to its section and repeat the drill rather than rereading passively.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.