Choosing the NDE Method by Damage Mechanism (API 510/570/653)

Short answer: pick the NDE method from the damage you expect, not from habit. API 510, API 570 and API 653 all tie examinations to credible damage mechanisms, using API RP 571 as the reference for how each one appears. General thinning calls for UT thickness or scanning. Localized corrosion and pitting call for area coverage: corrosion mapping, profile radiography, or MFL on tank floors. Surface cracking calls for WFMT, ACFM or ET. Embedded or sub-surface cracking calls for angle-beam UT, PAUT or TOFD.

The details are where inspection plans go wrong. A method that is excellent for one mechanism can be blind to another on the same component. Thickness readings at fixed points will not find chloride stress corrosion cracking. Wet fluorescent magnetic particle testing will not size internal HIC. Tank-floor MFL screens thickness loss but does not tell you which side the loss is on unless you check. This guide gives a mechanism-to-method matrix for vessels, piping and tanks. It explains what each method finds and misses, and how the codes expect the choice to be justified and recorded. It is a technique-selection guide. Setting inspection intervals by risk is a separate engineering exercise, covered only briefly here.

Scoping a turnaround or tank outage? Send us your damage mechanism list for an NDE scope and quote, answered within 24 hours.

What the codes require: method must match mechanism

The three in-service codes say the same thing in different words. NDE has to be chosen to find the damage that is credible for the equipment.

• API 510 (pressure vessels, 11th edition, 2022). Every vessel and pressure-relieving device in scope needs an inspection plan covering all credible damage mechanisms. A corrosion specialist is consulted to identify those mechanisms and the areas susceptible to localized corrosion, cracking, CUI and metallurgical damage. The methods and extent of NDE must be evaluated to make sure they can identify the mechanism and the extent and severity of the damage. Intervals consider, among other things, the type of mechanism, its rate, the equipment's tolerance to it, and the probability that the NDE method will detect it.
• API 570 (piping, 5th edition, February 2024). Piping systems are evaluated for present or possible damage mechanisms, and the NDE methods and extent must be able to identify the mechanism and its severity. Inspectors are expected to know the damage types in API RP 571 and the practices in API RP 574. Circuits are usually defined by grouping piping with common mechanisms, materials and conditions.
• API 653 (tanks, 5th edition through Addendum 3). When setting subsequent internal intervals from corrosion rates, the owner/operator should understand how effective the techniques used are at detecting and measuring the potential damage mechanisms. The standard also notes that bottom examination methods differ in how reliably they measure general corrosion and pitting.

The common reference is API RP 571, Damage Mechanisms Affecting Fixed Equipment in the Refining Industry. The current edition was published in 2020. For each mechanism it describes the affected materials, critical factors, affected units, appearance, prevention, and inspection and monitoring. API 571 tells you what to look for. Choosing, qualifying and applying the examination technique is NDE engineering, and that is where an ASNT Level III adds value.

Mechanism-to-method matrix

The table groups common mechanisms into families. It lists the methods usually first in line and the main limitation to plan around. It is a starting point for a written plan, not a substitute for the mechanism-specific guidance in API RP 571, the owner's procedures, or a qualified Level III's review of the specific geometry and material.

For more on individual mechanisms, see our field guide to API 571 damage mechanisms.

What each method finds and what it misses

Ultrasonic thickness (spot UT)

Spot UT measures remaining wall at a point and is the backbone of CML programmes. It is quick and repeatable when probe, couplant and locations are controlled. Its weakness is coverage. It finds what is under the probe and nothing between grid points. That is fine for general thinning and poor for pitting, erosion and CUI.

Corrosion mapping (automated UT and PAUT)

Mapping records thickness over an area, so it finds the minimum rather than sampling it. It suits localized corrosion, hot spots identified by spot UT, nozzle regions and baseline surveys that later surveys can be compared with. Coverage, index step and data review set its value, so specify them. See corrosion mapping.

Angle-beam UT, phased array and TOFD

These volumetric weld and plate techniques detect and size embedded flaws: HIC and SOHIC, weld cracking, lack of fusion, and fatigue cracks growing from the surface. PAUT gives imaging and coverage from fewer positions. TOFD is strong at through-wall height sizing. Both depend on procedure qualification, calibration and the operator's skill. See phased array inspection and TOFD testing.

Wet fluorescent magnetic particle (WFMT) and MT

WFMT is the standard surface crack method on carbon steel for wet H2S and other cracking surveys. It is very sensitive on clean, ground surfaces. It does not work on austenitic stainless steel, and it gives no depth information.

Penetrant testing (PT)

PT finds surface-breaking flaws on non-magnetic materials, including chloride SCC on stainless steel. It needs clean, dry, open cracks and gives no depth. Rough or contaminated surfaces reduce its sensitivity.

Eddy current and ACFM

Surface eddy current and ACFM can detect surface cracks through thin coatings, which reduces surface preparation. ACFM can also estimate crack length and depth. Tube eddy current techniques serve heat exchanger bundles. All depend on the right probe, frequency and reference standard for the material.

Radiography (profile, real-time and weld RT)

Profile radiography shows wall loss, deposits and erosion on small-bore and insulated piping without stripping insulation. Weld RT detects volumetric weld flaws. It brings radiation safety controls and licensing, and it is poor at finding tight planar cracks that are not aligned with the beam.

Guided wave, pulsed eddy current and MFL screening

These are screening tools. Guided wave covers long lengths of pipe from one location and flags areas of change for follow-up. Pulsed eddy current estimates average wall thickness through insulation. Tank-floor MFL screens plate area for metal loss above a threshold. None of them gives the final thickness. Each needs UT prove-up of the indications it raises. See MFL tank floor scanning.

NDT tank inspection: matching methods to tank damage

For aboveground storage tanks under API 653, the credible damage set is short and well known: soil-side floor corrosion, product-side floor pitting, shell thinning (often at the liquid line and lower courses), roof and structural corrosion, weld cracking in some services, and settlement. The usual method set follows from it:

• Floors: MFL scanning to the owner's threshold, UT prove-up of indications, pit gauging of visible product-side pits, and other methods for the critical zone near the shell, sumps and lap welds. Results feed the minimum remaining thickness calculation that sets the next internal interval.
• Shells: UT thickness surveys, which may be done from outside while the tank is in service, with mapping where localized loss is found.
• Welds: visual examination and MT or PT at repair locations and nozzle welds, and vacuum box testing of bottom welds after repairs, as the repair and examination sections of API 650 and API 653 require.
• Settlement: survey measurements evaluated under API 653 Annex B.

API 653 requires the owner's authorized inspector to assure the quality and completeness of NDE results at internal inspections, and to evaluate them. When you buy "NDT tank inspection", you are buying the examination data that inspector needs. See aboveground storage tank inspection.

Worked example, described qualitatively

Consider a carbon steel overhead drum and its outlet piping in a sour water service, plus a nearby crude tank. The plant's corrosion specialist lists the credible mechanisms. On the drum and piping: wet H2S cracking, ammonium bisulfide corrosion with erosion-corrosion at high-velocity locations, and CUI on insulated sections. On the tank: soil-side floor corrosion and product-side pitting at the water draw-off.

• Drum, wet H2S: WFMT on internal welds and attachment welds during the outage. Angle-beam UT or PAUT on welds and adjacent plate where HIC or SOHIC is credible. UT mapping of any blistered areas.
• Outlet piping, erosion-corrosion: UT mapping at elbows, reducers and downstream of control valves, rather than a single CML point per fitting. Profile RT on small-bore branches.
• Insulated sections, CUI: visual survey of the jacketing, screening of suspect areas, and insulation removal with UT at the locations the screening and visual flag.
• Tank: MFL floor scan at the owner's threshold, UT prove-up, pit gauging at the draw-off, and coverage of the critical zone.

Each line of the plan names the mechanism, the method, the extent, the procedure and the qualified examiner. The inspector of record then has a plan that answers the code question: can these techniques adequately identify this damage and its severity?

Inspection effectiveness and RBI: context only

Readers who use risk-based inspection will know the idea of inspection effectiveness. In the API RP 580 and 581 framework, an inspection is graded by how likely it is to detect and quantify a given damage mechanism, given the method and extent. The RBI assessment then uses that grade. It is a useful discipline even outside RBI, because it forces the question this page is about: will this method, at this coverage, find this damage?

Setting intervals with RBI is an engineering assessment the owner commissions under API RP 580/581 and the in-service code's RBI provisions. Atlantis does not perform RBI assessments. What we do is supply the technique detail the owner's RBI or inspection team needs: what each method detects, its coverage and its limitations. For background, see time-based vs risk-based intervals.

Personnel, procedures and documentation the inspector expects

• Written procedures for each technique, referencing the applicable ASME Section V article or owner specification, with calibration and reference standards defined.
• Qualified examiners. Personnel certified under the employer's written practice to ASNT SNT-TC-1A or an equivalent, at the level the work requires. Owners often require ASNT UT Level II for thickness work, and some owners add API QUTE for specific UT scopes.
• Coverage records: what was examined, with what technique, and what was not examined and why.
• Results tied to locations (CML numbers, drawing references, floor maps) so the next survey can be compared like for like.
• Indication reports classifying each finding by type and size, for the inspector to evaluate against the code or refer for engineering assessment.

Common mistakes in method selection

• Using spot UT for everything, including mechanisms that produce pitting, erosion or cracking.
• Relying on screening results (guided wave, PEC, MFL) as final thicknesses without prove-up.
• Using MT on austenitic stainless steel, or PT on rough surfaces without preparation.
• Specifying WFMT for HIC. It finds surface cracks, not mid-wall stepwise cracking.
• Copying last outage's scope after a process change or an integrity operating window exceedance that introduced a new mechanism.
• Not recording coverage, so later readers cannot tell "not found" from "not examined".
• Treating a mechanism list as finished. API 510 expects the plan to be reviewed and amended, including after management-of-change events.

Regulatory overlay: PSM, SPCC and Canada

Under OSHA PSM (29 CFR 1910.119(j)), inspection and testing of covered equipment must follow recognized and generally accepted good engineering practices, at frequencies consistent with them and with operating experience. Using API 510, 570 and 653 together with API RP 571 is the usual way plants show that. For tanks, EPA SPCC (40 CFR 112) requires integrity testing on a regular schedule according to industry standards. In Canada, provincial pressure equipment regulators (for example ABSA in Alberta and TSSA in Ontario) oversee owner inspection programmes, and many accept or reference the API codes. Confirm what your jurisdiction expects before finalizing the plan.

How Atlantis supports this

Atlantis NDT performs the examinations your inspection plan calls for: UT thickness and corrosion mapping, PAUT and TOFD, WFMT, MT, PT, eddy current, guided wave screening and MFL tank floor scanning. The work is done by ASNT-certified technicians under ASNT Level III oversight, and radiography is done by crews licensed where the work is. Our Level III reviews the damage mechanisms the owner's corrosion specialist has identified and recommends matching techniques, procedures and coverage. We then perform the NDE and report results with coverage maps for the owner's API inspector, who remains inspector of record and owns the plan. Atlantis does not perform RBI or fitness-for-service assessments. See inspection services and Level III consulting. Over 1,500 inspection activities completed.

Request an NDE scope review and quote.

Frequently asked questions

What is NDT tank inspection?

The nondestructive examinations used to assess an aboveground storage tank: MFL and UT on floors, UT thickness on shells, visual examination, and MT, PT and vacuum box testing on welds and repairs. Under API 653, the results feed the authorized inspector's evaluation and interval calculation.

What does "API NDT" mean?

Usually NDT performed to support API in-service inspection codes (API 510, 570, 653). The examinations are done by qualified NDE personnel, and the results are evaluated by the owner's API-certified inspector.

What should I look for in an API NDT company?

Written procedures for each technique, ASNT-certified examiners under a written practice, Level III oversight, the ability to cover the methods your damage mechanisms need, and reports that record coverage and locations. Also confirm the company does not claim to be your authorized inspector unless it actually holds that role.

Which NDT method is best for corrosion?

For general thinning, UT thickness. For localized corrosion and pitting, area methods such as UT or PAUT corrosion mapping. For screening large areas, MFL on tank floors and guided wave or pulsed eddy current on piping, followed by UT prove-up.

Which NDT method finds stress corrosion cracking?

PT on stainless steel, WFMT or ACFM on carbon steel, and eddy current through coatings for detection. Angle-beam UT or PAUT is then used to estimate depth.

Does API 571 tell you which NDE method to use?

It gives inspection and monitoring guidance for each mechanism. The final choice of technique, procedure and extent belongs in the owner's inspection plan, reviewed for the specific equipment.

Can one method cover all damage mechanisms?

No. Thickness methods do not find cracks, and surface crack methods do not measure wall loss or embedded flaws. Most equipment needs a combination.

What is inspection effectiveness?

A grading, used in API RP 580/581 risk-based inspection, of how well an inspection method and coverage detect a given damage mechanism. It is part of the owner's RBI work. Atlantis supplies the technique detail but does not perform RBI.

Who decides the NDE methods in an API inspection plan?

The inspector and/or engineer who develops the plan, with a corrosion specialist identifying mechanisms. An NDE Level III can advise on technique capability and procedures.

Do you perform NDT tank inspections outside Texas?

Yes. Atlantis performs onsite NDE at the client's facility. Contact us with the tank location and scope for a quote.

Planning NDE by damage mechanism? Get a quote from our Level III team.

Related reading: damage mechanism screening per API 571 and tank floor pitting and MRT.