Magnetic Particle Testing (MT) — Complete Field Guide

How the physics works

Magnetic particle testing exploits the fact that a crack in ferromagnetic material distorts an applied magnetic field, forcing flux to leak out of the surface at the discontinuity. Fine ferromagnetic particles — dry powder or wet suspension — applied to the surface migrate into the leakage field and concentrate over the defect, making it visible to the naked eye or under UV-A for fluorescent particles. ASME BPVC Section V Article 7 sets the procedure rules [1]; ASTM E709 covers practice [2]; ASTM E1444 covers aerospace MT [3]. The magnetic field must be roughly perpendicular to the suspected crack orientation — a longitudinal field detects transverse cracks, a circular field detects longitudinal cracks. Two orthogonal magnetizations per area are mandatory under most codes because no single field direction catches all crack orientations.

When to choose this method

Choose MT when the material is ferromagnetic, the suspected defects are surface or near-surface (up to about 0.060" depth for dry-powder yoke work), the geometry permits two-orthogonal magnetization passes, and the part can tolerate residual magnetism. MT is the standard for in-process and final weld inspection on carbon steel structural and pressure welds, and it is the highest-sensitivity surface method on ferromagnetic material — wet fluorescent MT under UV-A in a dark inspection booth resolves indications PT cannot match.

MT cannot find defects deeper than roughly 1/16" below the surface; volumetric or subsurface flaws need UT or RT. It does not work on non-ferromagnetic material — austenitic stainless (304, 316), aluminum, copper, titanium, nickel-base alloys. Heavy coatings (>0.005") suppress the leakage field. Residual magnetism is unacceptable on parts that will operate near sensitive instruments — instrumentation skids, MRI components — and demagnetization is mandatory before release. Field-induced arc burns from prod-method MT can damage high-strength steel and are prohibited on aerospace and pressure-vessel work.

Materials & geometries

Method coverage depends on couplant, surface, and section thickness. Compatible forms include:

Procedure

Procedure qualification and surface prep

A written procedure references the governing code (ASME V Article 7, AWS D1.1 §6.14, API 1104 §11.5) and specifies the magnetizing method (yoke, prod, central conductor, coil), the particle type (dry vs wet, visible vs fluorescent), the field strength verification (Pie gauge or QQI shim), the lighting (≥100 footcandles ambient for visible, ≥1000 µW/cm² UV-A for fluorescent), and the acceptance reference. The procedure is reviewed and signed by a Level III qualified to ASNT CP-189 [4] or SNT-TC-1A.

Surface prep removes loose scale, slag, weld spatter, paint flakes, and oil that would mask indications or block particle mobility. ASTM E709 §7 [2] requires the surface to be dry and free of contaminants that would prevent particles from migrating freely. A weld inspected for hydrogen cracking sits 24-48 hours after welding before MT — delayed cracks need time to develop, and inspecting too soon misses them entirely. AWS D1.5 §6.14 explicitly mandates a delay for HSLA bridge welds [5].

Magnetization technique selection

The yoke method dominates field work — Magnaflux Y-7 or Parker B-310 AC/DC electromagnetic yoke, 10-pound lifting test on AC or 40-pound on DC verifies adequate field strength under ASTM E709 §8.1.2 [2]. AC yokes give the best surface-crack sensitivity because skin effect concentrates flux at the surface. DC yokes penetrate slightly deeper (0.060" vs 0.030" practical) and catch shallow subsurface defects AC will miss. Two perpendicular yoke positions per inspection area are mandatory — typically along the weld axis and across it.

Prod and central-conductor methods are used in shop work on larger parts. A prod pair injects current directly into the part at 100-125 amps per inch of prod spacing under ASTM E709 §8.4. Prods are banned on most aerospace and high-strength steel work because of arc-burn risk; central conductor through a hollow shaft or bolt avoids the burn problem entirely. Coil-shot magnetization (placing the part inside a 5-turn coil) gives a longitudinal field for inspecting transverse cracks on smaller parts.

Particle application and field interaction

Dry powder (Magnaflux 8A red, 1 gray, 3A yellow) is dusted onto the surface during magnetization using a powder bulb or shaker. Excess powder is removed with a light air bulb stream, never wiped. The technician watches for particle migration during application — the indication forms while the field is active, not after. Continuous-method MT applies particles during the magnetizing pulse and holds the pulse until the indication develops, which is the dominant field technique today.

Wet fluorescent MT (Magnaflux 14A or Met-L-Check WCP-2) suspends the particles in a kerosene or water carrier and applies them via spray gun before or during the field. Under UV-A at ≥1000 µW/cm² in a darkened booth (<2 footcandles ambient white light), the indication glows yellow-green. ASTM E1444 §10 [3] mandates the lighting check at the start of every shift with a calibrated UV radiometer (Spectroline DSE-100X).

Field strength verification

The Pie gauge (a notched disc of low-carbon steel) and the QQI (quantitative quality indicator) shim verify that the field at the inspection surface is adequate. ASME V T-754 [1] requires a verification of field direction and adequacy at the start of every inspection and any time the equipment, geometry, or technique changes. A QQI shim shows a clear indication on the engraved notches when the field at the surface meets the minimum 30 oersteds (2400 A/m) typically called out by aerospace MT procedures.

Gauss meters with Hall-effect probes (F.W. Bell 5180, Lake Shore 410) measure residual field after demagnetization. Most pressure-equipment specs cap residual at 3 gauss; aerospace and instrumented assemblies often cap at 1 gauss. Residual above the cap requires demagnetization through a reversing DC coil or an AC reverse-step procedure.

Indication interpretation and disposition

Indications fall into three categories: relevant (real discontinuities), non-relevant (geometric — keyways, splines, magnetic-write across mating surfaces), and false (loose particles, contamination). The technician distinguishes them by re-magnetizing perpendicular to the suspected indication, cleaning and re-particling, and confirming the indication reappears in the same location with the same orientation. Non-relevant indications get noted on the report; false indications are cleaned and disregarded.

Acceptance varies by code. AWS D1.1 §6.14 [5] accepts no cracks, no incomplete penetration or fusion at the surface, and limits porosity per the visual table. ASME B31.3 §344.3.2 accepts MT to the workmanship limits of Table 341.3.2 [6]. API 1104 §11.5 [7] accepts to its own surface-defect table for pipeline welds. Any crack indication is rejectable regardless of length under every common code.

Demagnetization and reporting

Residual magnetism interferes with subsequent welding (arc blow), machining (chip adhesion), and final assembly. AC demagnetization through a reversing coil at gradually decreasing current is the field standard. The gauss-meter check after demag confirms the part meets the spec residual limit. For large weldments, a low-frequency reversing DC field through a coil wrapped around the part demagnetizes more uniformly than an AC field that may not penetrate the thickness.

The MT report names the procedure, magnetization method, current type and amperage, particle type and color, lighting (visible or UV-A intensity), field verification method, all reportable indications with location and dimension, and demagnetization confirmation. ASME V §T-790 [1] lists every required field. Digital reporting systems (Atlantis NDT Reporting, Sonatest CapturePro) auto-populate the form from a barcode of the procedure and a photo of the indication.

Equipment

Yokes and current sources

The Magnaflux Y-7 and Parker B-310 are the dominant field-portable AC/DC yokes in North American service. Both deliver the ASTM E709 [2] 10-lb AC and 40-lb DC lifting test at full field. A spare yoke and a spare cord ride in every MT crew kit because cord failure mid-shift is the most common downtime cause.

For shop work, mobile units like the Magnaflux Multi-MX and Parker DA-750 deliver 5,000-A half-wave DC for thick-section bench inspection. Pulse-method circular magnetization through clamps gives high-current short-duration shots that minimize part heating — useful on large castings where prolonged current would burn the contact pads.

Particles and carriers

Magnaflux 8A red, 1 gray, 3A yellow dry powders cover the visible-light field market. Magnaflux 14A fluorescent wet suspension dominates shop wet-fluorescent MT. Carrier choice between kerosene-based and water-based depends on the part — water-based is required on potable-water-service equipment and many food-processing applications, kerosene-based gives better wetting on oily surfaces and is the default for shop work.

Particle concentration is verified daily with a centrifuge tube (Ascarite or Erlenmeyer 100 ml graduated tube) — typically 0.1 to 0.4 ml settled volume per 100 ml of suspension for fluorescent particles. ASTM E709 §X1.6 [2] gives the method and the acceptance range.

Lighting and field verification gear

For visible MT, a white-light meter (Spectroline DM-365XL) verifies ≥100 footcandles at the inspection surface. For fluorescent MT, a UV-A radiometer (Spectroline DSE-100X) verifies ≥1000 µW/cm² UV-A and <2 footcandles ambient white light. UV-A lamp bulbs degrade with age — output check at the start of every shift, replacement at any reading below the procedure threshold.

Pie gauges (ASTM E709 Figure 11) and QQI shims (Magnaflux 3/8 or 1/2 inch sizes) verify field direction and adequacy at the part surface. A gauss meter (F.W. Bell 5180) measures residual field after demagnetization. All field-verification gear is calibrated annually against NIST-traceable standards.

Codes & standards that govern this method

Procedures and acceptance criteria are anchored in published codes:

Acceptance criteria

For new-construction structural welds under AWS D1.1 §6.14 [8], any crack indication is rejectable regardless of length, lack-of-fusion indications at the surface are rejectable, and porosity follows the visual acceptance table. For ASME B31.3 piping welds, MT acceptance is per Table 341.3.2 — no cracks, no incomplete fusion, linear indications limited to 3/16" for thickness >3/4" [6]. API 1104 §11.5 [7] applies to pipeline construction and adds specific limits for arc burns and undercut measurable through MT. In-service MT under API 510 and API 570 follows the original construction code acceptance for the welded joint, modified by any approved repair-and-rerate procedure. Hydrogen-assisted cold cracks are delayed indications that may not appear for 24-48 hours after weld cooldown — MT performed too early misses them, and inspection timing must follow the procedure delay even when the schedule pressures push for immediate sign-off.

How this compares to other methods

Choosing between methods is rarely about capability alone — cost, throughput, and code coverage all weigh in:

Cost range

Typical MT pricing in the US market runs $2–$6 USD per linear foot of weld inspected (Gulf Coast crew, 2025), with most jobs landing around $3.5 USD per linear foot of weld inspected (Gulf Coast crew, 2025). Mobilisation, access, and certification level shift the band.