Magnetic Flux Leakage vs. Eddy Current for a Ferromagnetic Storage Tank Floor
A terminal operator asks you to select a method for inspecting the floor of an above-ground storage tank. The floor plates are 6 mm carbon steel (relative permeability ~200-500, conductivity ~7 %IACS). The service history includes water bottoms and possible underside corrosion. The operator's draft scope says: 'Use eddy current because it is fast and does not need couplant.' The tank is 30 m diameter, the floor is lap-welded, and the operator wants detection of 20 % wall loss (about 1.2 mm). The floor is coated on the product side with a 300 µm epoxy, and there is no access to the underside.
1. Evaluate the proposed eddy current approach for this tank floor. Is it appropriate? What method would you recommend?
Read the worked answer
Conventional eddy current is a poor choice for 6 mm carbon steel. The material is ferromagnetic (relative permeability in the hundreds), which makes the eddy current response dominated by permeability variations rather than by wall loss, and the standard depth of penetration at practical frequencies is a fraction of a millimeter, far short of 6 mm. Eddy current also cannot penetrate the 300 µm epoxy coating meaningfully for a through-wall assessment.
The appropriate method is magnetic flux leakage (MFL), which magnetizes the ferromagnetic floor and detects flux leakage at wall-loss regions, or a low-frequency electromagnetic technique such as remote-field or pulsed eddy current if a non-contact, through-coating assessment is desired. For a 6 mm floor with a 1.2 mm detection target, MFL is the standard shop-floor method; specify a magnetizer sized to saturate the plate and a reference plate with machined flat-bottom holes at 20 % and 40 % of wall. Note that MFL gives a relative indication, not a direct wall-thickness measurement, so any flagged region must be verified by ultrasonic thickness gauging.
2. The operator insists on a through-coating technique. What are your options, and what are their limitations?
Read the worked answer
Options: (1) Pulsed eddy current (PEC) can inspect through coatings and insulation and gives a relative wall-loss estimate; it is sensitive to lift-off but can be normalized. Its limitation is that it averages over a footprint of tens of millimeters, so small pits may be missed, and it requires a reference standard on the same material. (2) Low-frequency eddy current with a suitable probe can work through thin coatings but is limited by the ferromagnetic permeability noise and shallow penetration on 6 mm steel. (3) MFL with a coated floor is still viable if the magnetizer can saturate through the coating; the coating thickness must be accounted for in lift-off.
Recommendation: PEC for a fast through-coating scan, with MFL or UT verification of flagged areas. Document that no single technique gives both full coverage and direct thickness on a coated ferromagnetic floor, and that verification by UT is mandatory before repair decisions.
3. What does the 20 % wall-loss detection target imply for the reference standard and the acceptance criteria?
Read the worked answer
The reference standard must contain machined flat-bottom holes or EDM slots representing 20 % and 40 % of the 6 mm wall (1.2 mm and 2.4 mm deep) in the same carbon steel, so the technique can be demonstrated to detect the 20 % target at an acceptable signal-to-noise ratio (typically ≥3:1). Acceptance criteria should be written against the demonstrated capability: indications at or above the 20 % reference are flagged for UT verification, and indications at or above 40 % are candidates for repair. The Level III must state that detection is probabilistic and that the demonstration on the reference standard defines the minimum detectable loss, not a guarantee for every pit.