ASME Section V Article 7 — Magnetic Particle Examination

What Article 7 covers, and what it deliberately does not

Article 7 of ASME Section V is the method article for magnetic particle examination of ferromagnetic material. It defines the magnetising techniques the Code recognises, namely the yoke, prods, the encircling coil and the central conductor, together with dry and wet particle systems in both visible and fluorescent form, the surface condition required, the lighting under which indications are viewed, verification of field adequacy, and demagnetisation. Article 1 sits above it and supplies the general requirements: a written procedure, personnel qualified under the employer's written practice, and record retention.

What Article 7 does not do is decide anything. It contains no accept or reject limits, no coverage rules for a given weld joint, no examination extent, and no timing within the fabrication sequence. Every one of those comes from the referencing code section or the purchase specification. An MT procedure that cites only Article 7 and stops is incomplete, and an inspector who tells you a linear indication is rejectable per Article 7 has reached for the wrong document.

The material limit is absolute rather than a matter of judgement. Magnetic particle examination works only where the substrate carries a magnetic field, so austenitic stainless steel, aluminium, copper alloys, titanium and most nickel alloys sit outside the method entirely. Where a ferritic base metal has been welded with austenitic filler, the weld metal itself may not respond even though the parent plate does, and the correct call is liquid penetrant to Article 6. Getting that decision wrong at procedure stage costs a full re-examination later, and a review by ASNT Level III consulting before the procedure is issued is cheaper than remobilising to a fabricator.

Choosing between yoke, prods, coil and central conductor

The yoke is the default for weld examination in the field because it is portable, leaves no contact damage, and produces a longitudinal field between its poles. Its weakness is area and depth. The useful field lies broadly between the legs, so coverage is built from many overlapping placements, and each placement has to be rotated to catch indications in the second direction. Alternating current concentrates the field at the surface and is the better choice for surface-breaking cracks; half-wave and full-wave direct current penetrate further and suit near-surface discontinuities within a shallow band.

Prods pass current directly through the part and generate a circular field around the current path. They cover ground faster than a yoke on large plate, but they carry a real risk: the contact points arc. Article 7 requires contact pads, and any arc burn has to be removed and the area re-examined, because a burn is a hard, untempered spot and a stress riser in service. Many owner specifications for pressure equipment simply prohibit prods on finished surfaces, and aerospace work bans them outright. That constraint belongs in the technique sheet before the first shot, not in a concession request afterwards.

Coils and central conductors belong in the shop. An encircling coil produces a longitudinal field with a useful length of roughly six to nine inches either side of the coil, so long shafts are examined in overlapping shots, and the fill factor of the part inside the coil changes the effective field strength. A central conductor threaded through a nozzle forging, a ring or a length of tube produces a circular field with no electrical contact on the part at all, which makes it the technique of choice wherever prod burns would be unacceptable.

The two-direction rule, and why single-shot MT fails review

Magnetic particle examination only reveals discontinuities that interrupt the magnetic field. An indication lying parallel to the flux leaks almost nothing and can be missed entirely, while the same discontinuity oriented at 90 degrees to the field produces a strong, sharply defined build-up of particles. Article 7 therefore requires two examinations of each area with the field directions approximately 90 degrees apart. This is a requirement, not a recommendation to be traded away against schedule pressure.

In practice the failure is procedural rather than technical. A technician places the yoke along the weld axis, walks the length of the joint, finds nothing, and signs the report. Transverse cracking, the population that matters most in a hardenable material inside the hydrogen-delayed cracking window, was never sought, because the field never ran across the weld. The report looks complete. It is not, and any reviewer comparing the recorded field directions against the sketch will see the gap immediately.

The durable fix is a report format that forces both field orientations to be recorded per area, with the overlap stated, so that an incomplete record cannot be closed out. That is a documentation control rather than a technical one, and teams running magnetic particle work across several fabricators usually push it into their inspection management software so the constraint sits in the system instead of in a technician's memory.

Field adequacy: lifting force, Hall probe and field indicators

Article 7 does not accept the assertion that the yoke was working. Field adequacy has to be demonstrated. The classic demonstration is lifting force, 10 lb with alternating current and 40 lb with direct current, and the sting is in one detail that gets skipped: the weight must be lifted at the maximum pole spacing that will actually be used on the job. A yoke that comfortably lifts 10 lb with the legs closed to four inches can be well short of adequate at eight inches, which is exactly where it will be used to reach around a nozzle.

A Hall-effect tangential field gauge measures the field itself, generally expected in the 30 to 60 gauss band at the surface, and it is the only method that yields a number tied to the part rather than to the equipment. It has to be within calibration and the probe has to sit flat against the surface in the orientation the standard assumes. Pie-shaped field indicators and flexible strip indicators demonstrate that a field exists and roughly where it runs, which is useful for confirming direction and overlap, but they do not measure strength and should never be presented in a report as though they did.

Verification is required at defined moments: at the start of the examination, after any change in an essential variable, after replenishment of a wet bath, and at the end of the examination. The end-of-job check is the one crews skip, and it is the one that carries the most weight. If it fails, everything examined since the last good verification is in question, which is precisely why the requirement exists and precisely why an auditor asks for it first.

Particles, bath concentration and viewing conditions

Dry powder is applied while the current is on and the excess removed with a gentle air stream. Gentle matters, because a hard blow-off strips the indication along with the background. Dry powder tolerates rough and warm surfaces better than a wet suspension, which makes it the usual pairing with a yoke on as-welded surfaces in the field, where the alternative is trying to keep a bath clean on a scaffold.

Wet suspensions carry a measurement that is routinely neglected: bath concentration, checked in a centrifuge tube and typically held between 0.1 and 0.4 mL of settled solids per 100 mL for fluorescent baths. A tank topped up with carrier alone drifts lean and quietly stops forming indications, while an over-concentrated bath fluoresces across the whole surface and masks them. The check takes a few minutes at shift start, and along with the contamination check on the carrier it is among the first records requested in any review of a wet bench.

Viewing conditions carry hard numbers. Fluorescent examination requires at least 1000 microwatt per square centimetre of UV-A at the surface being examined, ambient white light held to about 2 foot-candles, and a dark adaptation period before the inspector begins calling indications. Visible colour-contrast examination requires a minimum of 100 foot-candles of white light. The recurring error is measuring intensity at a convenient bench distance and then working at arm's length inside a vessel, where the intensity at the surface has fallen by a large factor. Technicians who understand why the reading is taken at the surface rather than at the lamp generally learned it in structured NDT training rather than on the job.

Demagnetisation and the residual field threshold

Demagnetisation is not universally required, and procedures that demand it everywhere waste time. Article 7 calls for it where residual magnetism would interfere with what happens next: further welding, subsequent machining where ferrous swarf clings to a finished surface, or service in instrumentation, bearings or rotating assemblies where a residual field is a nuisance or a hazard.

The dominant case in pressure work is further welding, because residual field causes arc blow. The welder sees the arc pull away from the joint, the deposit undercuts on one side, and porosity or lack of fusion follows. It is usually blamed on the consumable or on the welder long before anyone reaches for a field indicator. Where magnetic particle examination has been performed on a joint that will receive more passes, demagnetise and then measure. Three gauss or less on a tangential field indicator is the accepted working threshold, and the measured value belongs on the report rather than in a tick box.

The method matters as much as the result. Alternating current demagnetisation with decaying amplitude works well on thin sections but penetrates poorly in heavy wall, where reversing and decreasing direct current is what a forging actually needs. A procedure that says demagnetise as required, with no method, no equipment and no measured limit, is a finding waiting to be written against it.

Where the acceptance criteria actually come from

For a Section VIII Division 1 vessel, magnetic particle acceptance sits in Mandatory Appendix 6: relevant indications are evaluated above a defined length, cracks and linear indications are unacceptable regardless of length, and rounded indications carry their own limits. For process piping to B31.3 the criteria live in the tables that correspond to the applicable fluid service. For structural steel to AWS D1.1 the clause depends on whether the connection is statically or cyclically loaded. None of those numbers appear anywhere in Article 7.

The consequence is that an MT procedure has two halves with different owners. Article 7 governs the technique. The construction code or the purchase specification governs what is rejectable. Where a client specification is more restrictive than the code, which in refinery and offshore work it usually is, the procedure has to name the document the acceptance level comes from, by clause. Procedures that say acceptance per applicable code are returned, because they leave the interpretation to whoever happens to be holding the report at the time.

This split is also where independent review earns its keep. A batch of magnetic particle reports can be flawlessly executed to Article 7 and still be evaluated against the wrong acceptance table, and that only becomes visible when someone reads the reports against the purchase specification rather than against the method. That, in substance, is what independent NDT report validation does to a fabrication package before it is accepted.

Findings that recur, and the traps that catch buyers

The findings repeat across fabricators with unhelpful consistency. Only one field direction recorded. Field adequacy verified at the start of the job and never at the end. A Hall gauge whose calibration lapsed part way through a contract. Bath concentration checks missing for entire weeks. UV-A intensity measured, but not at the surface. Prod burns left in place. Demagnetisation claimed and never measured. And a procedure whose acceptance criteria have been lifted out of the method article instead of the construction code.

Two commercial traps are worth naming plainly. First, an approved procedure is not an approved technique. The procedure permits a range of essential variables, and the technique actually applied to your joint has to sit inside that range, which is only visible on the report and the technique sheet. Second, personnel qualification runs to the employer's written practice rather than to a certificate viewed in isolation. The written practice, the examination records behind it, and the current vision test all have to exist for the specific individual working on your equipment.

Both are answerable before mobilisation instead of during an audit. A documented review of the written practice, the procedures and the technique sheets against the purchase specification is a short exercise with a long payoff, and it is the usual starting point for Level III technical authority support on a new fabrication contract or a first-time supplier.

Frequently asked

Is ASME Section V Article 7 the same as ASTM E709?

No. E709 is a standard guide for magnetic particle testing, adopted by ASME as SE-709, and it provides detail and background that Article 7 references rather than repeats. Article 7 is the mandatory Code article. Where a purchase order invokes ASME Section V, the procedure is written to Article 7 and uses SE-709 for supporting technique detail. Citing only E709 on a Code item leaves the mandatory requirements unaddressed.

Can Article 7 magnetic particle examination be used on stainless steel?

Only on ferritic and martensitic grades that respond to a magnetic field. Austenitic grades such as 304 and 316 are not ferromagnetic in the annealed condition and cannot be examined by magnetic particle at all, whatever the equipment. Duplex grades respond, but ferrite content and surface condition affect sensitivity and the technique needs to be demonstrated. Where the material will not support a field, the correct surface method is liquid penetrant to ASME Section V Article 6.

How often does yoke lifting force have to be verified?

At the start of the work and whenever the yoke is repaired, with the check performed at the maximum pole spacing that will be used. Many owner specifications add a daily or per-shift check and a formal annual verification with a certified weight. The point is not the frequency in isolation but the exposure: if a check fails, all work back to the previous successful check has to be reconsidered, so a longer interval means a larger loss.

Does Article 7 require the residual field to be measured after demagnetisation?

Where demagnetisation is required, the result has to be demonstrated rather than assumed, and a measured residual field on a tangential field indicator is the practical way to do it. Three gauss is the usual working limit unless the referencing code or specification is tighter. Recording a numerical value costs nothing at the time and settles the argument later when a welder reports arc blow on the same joint.

What is the most common Article 7 finding on a fabrication audit?

A single direction of magnetisation. It is the easiest requirement to skip because the work still looks complete, the report still carries a result, and nothing visibly goes wrong until a transverse crack is found downstream by another method. The second most common is field adequacy verification recorded at the start of the job but missing at the end, which leaves the whole examination with no closing demonstration that the equipment still performed.