steel-structure-corrosion-inspection
Steel Structure Corrosion Inspection: NDT, DFT & Corrosion Rate
A reflector-vested inspector presses an ultrasonic thickness probe—yellow coupling gel visible on the spot—against a steel column web, while recording readings on a clipboard. Behind him, steel beams, columns and a bolted joint frame an operating plant interior; the mood is objective, instrument-based measurement rather than a visual walk-around.
A paint chip is not a diagnosis. A flaking coating tells you something is wrong, but it does not tell you how much steel is already gone, whether a crack is propagating beneath the surface, or whether the problem is getting worse this year than last. A proper steel structure corrosion inspection uses non-destructive testing (NDT) to measure what the eye cannot: wall loss, surface cracks, coating thickness, and the rate at which corrosion is advancing.
This guide explains the three core NDT methods (UT, MT, PT), dry-film thickness (DFT) measurement of the coating, and how to convert wall-loss trends into a corrosion rate in mpy—then ties it all to inspection frequency and a report you can actually act on. Choosing a coating system at design time is covered in our steel structure corrosion protection article; deciding when to repaint is covered in our maintenance schedule guide. This one is about how to measure corrosion objectively.
Why Corrosion Inspection Is Not the Same as Coating Selection
Three different jobs get confused under the word "corrosion," and separating them is the first step to a useful inspection. Coating selection is a design-time choice: which primer, topcoat, galvanizing, or ISO 12944 exposure category (C3, C4, C5) the building gets new. Maintenance scheduling is an operations decision: how often to walk the building, when to repaint, when to do a full re-survey. Corrosion inspection—the subject here—is the diagnostic event: using NDT to measure the current condition and rate, and producing data that decides whether to repair now, derate, or keep monitoring.
The three work together. Selection sets the initial protection; inspection verifies whether that protection is actually holding up; scheduling sets how often the inspection repeats. An inspection is triggered at several points: at handover (factory NDT and site acceptance), on a periodic operating cycle (annual, biennial, or five-yearly depending on environment), after a flood or fire, or whenever there is reason to suspect deterioration. Independent verification is often the most defensible route—see steel building third-party inspection for how that works. And the cadence of when to look is detailed in steel structure corrosion maintenance schedule.
NDT Methods — UT, MT, PT
In any steel structure corrosion inspection, ultrasonic thickness testing (UT) fires high-frequency sound through the steel and times the echo from the back wall to read remaining thickness. It measures internal or hidden wall loss—inside a column web, a truss chord, or a pipe—from one accessible side, no need to reach both faces. Typical accuracy is about ±0.1 mm (±0.004 in) across a measurement range of roughly 1.5–200 mm (1/16–8 in). Its limits: the surface must couple cleanly (local rust or coating is ground back at the spot), and UT reads thickness, not crack orientation.
Magnetic particle testing (MT) magnetizes the part; at a surface or near-surface crack the field leaks and holds magnetic powder, drawing the flaw as a visible line. MT is the standard method for surface cracks on carbon-steel welds and fatigue-prone details—by far the most-used surface crack tool in structural steel. It works only on ferromagnetic steel (carbon and low-alloy), not on austenitic stainless, and it sees only surface and very near-surface flaws. The coating is ground back to bright metal before testing, then restored afterward and the part demagnetized.
Dye penetrant testing (PT) draws colored penetrant into an open crack and pulls it back out with developer, showing the flaw as a red line. It is used on non-ferrous welds and castings where MT cannot magnetize, and as a low-cost, portable surface screen. PT reads only surface-open flaws—it cannot see internal or subsurface cracks—and it does not work on porous material. Compared with MT, PT is cheaper and universal on material but less sensitive; MT reaches near-surface sub-surface cracks that PT misses.
The selection logic is straightforward: internal wall loss under paint → UT; surface cracks on carbon-steel welds → MT; surface-open flaws or quick non-ferrous screening → PT. Critical joints get combined UT + MT. These methods are applied to welds that were made to a controlled procedure—see steel building welding process—and on details designed to be inspectable; see steel structure connection design. The baseline factory NDT step is described in steel structure quality inspection.
| NDT Method | What It Detects | Material | Metric Accuracy | Imperial Accuracy | Cost Level |
|---|---|---|---|---|---|
| Ultrasonic (UT) | Internal wall loss, thickness | Any steel | ±0.1 mm | ±0.004 in | Medium–high |
| Magnetic particle (MT) | Surface & near-surface cracks | Ferrous only | Visual / mm grading | Visual / inch grading | Medium |
| Dye penetrant (PT) | Surface-open flaws | Any, non-porous | Visual only | Visual only | Low |
| Visual + tap | Surface rust, delamination | Any | Qualitative | Qualitative | Low |
Typical performance; exact tolerances follow equipment and procedure specs. consult our engineers for the right method grid on your members.
Coating Thickness (DFT) & Hidden Corrosion
Coating measurement is the second pillar of a steel structure corrosion inspection: the coating itself is measured with a dry-film thickness (DFT) gauge—magnetic-induction type on ferrous steel, eddy-current on non-ferrous. Good practice takes at least 5 readings per 10 m² (~100 sq ft), with extra points at high-risk zones (column bases, eaves, splash zones). The acceptance rule from ISO 12944 / ISO 19840 is a common one: at least 80% of spot readings must meet the design DFT, and no reading should exceed 200% of it—over-thick film cracks and delaminates just as under-thick film fails early. Unit conversion is simple: 1 mil = 25.4 μm, and a typical epoxy system runs 120–250 μm (4.7–9.8 mil) total DFT.
The trap is hidden corrosion under an intact coating. A blister or a single rust spot can mask widespread thinning underneath, because the coating looks fine from the outside. When DFT readings jump oddly, the response is to follow up with UT wall thickness at that spot; an ultrasonic A-scan separates the coating echo from the steel back-wall echo, so the gauge can read real remaining steel through good paint. Beyond thickness, a complete coating condition survey also deploys adhesion pull-off tests, cross-hatch grading, and holiday (spark) detection to find pinholes the gauge cannot—our dedicated guide to steel coating inspection and testing covers the full battery, from wet-film measurement on the wet line to pull-off adhesion on the cured film. The classic hiding places are the inner face of trusses and purlins, column-to-beam joints, and low spots where water pools. The coating system being measured is detailed in steel structure painting, and roof drainage (which sets where water ponding hides corrosion) is covered in steel building roof system. For the most hidden corrosion of all—behind insulation and jacketing on process piping and heated columns—our guide to steel corrosion under insulation explains the CUI mechanism, the 0–175°C temperature window, and the Pulsed Eddy Current NDT methods that detect it without stripping the cladding.
| Exposure Category (ISO 12944) | Typical Total DFT | DFT in Mils | Typical Environment |
|---|---|---|---|
| C2 — low | 120 μm | ~4.7 mils | Dry, inland, clean |
| C3 — medium | 160 μm | ~6.3 mils | Urban / industrial, moderate |
| C4 — high | 200 μm | ~7.9 mils | Coastal / chemical, moderate |
| C5 — very high | 250 μm | ~9.8 mils | Coastal splash / heavy industrial |
Illustrative total dry-film ranges; design DFT is set by the coating specification and environment. Verify against ISO 12944.
Need Objective Corrosion Data, Not a Visual Guess?
A paint chip on a column does not tell you whether the web has lost 10% or 30% of its thickness. We arrange third-party NDT—UT wall-thickness grids, MT on welds, DFT spot checks—with a written report you can hand to your structural engineer and your insurer.
Corrosion Rate Measurement
Reading that trend is what turns raw UT data into the decision output of a steel structure corrosion inspection. A single wall-thickness number is useful; the trend is the real story. Corrosion rate is expressed in mpy (mils per year) in US practice, converting as 1 mpy ≈ 0.0254 mm/year (or 1 mm/year ≈ 39.4 mpy). The simplest calculation is comparative: take this UT wall reading minus the previous UT reading, then divide by the years between surveys. Two more methods sharpen it: an embedded electrical-resistance (ER) probe reads real-time rate, and coupon (weight-loss) testing exposes standard specimens and weighs them over time. Unit and classification guidance is gathered under AMPP (NACE) Corrosion Standards.
Reading the rate tells you what to do. Below about 5 mpy (0.13 mm/yr) is low—keep the existing plan. 5–20 mpy (0.13–0.5 mm/yr) is moderate—shorten the inspection interval and patch locally. Above 20 mpy (> 0.5 mm/yr) is aggressive—run a full remaining-life evaluation and consider derating or strengthening. Remaining life is a simple ratio: t = (original thickness − minimum allowed thickness) ÷ annual corrosion rate, which is exactly why the trend line matters more than any one number. Continuous monitoring between surveys is increasingly practical—see steel structure IoT monitoring—and after a flood or splash event, a re-survey can reset the rate upward; see steel building post-disaster assessment. After a flood or splash event, a post-flood corrosion re-survey can reset the rate upward sharply—floodwater brings chloride into contact with steel that has been dry for years, accelerating corrosion 3–5× at column bases and hidden nodes. The same UT wall-thickness grid that tracks normal aging becomes the tool that sizes the flood repair: wash, blast to Sa2.5, recoat, or replace depending on section loss.
Beyond periodic UT surveys and environmental IoT sensors, corrosion rate data can feed directly into a continuous structural monitoring system that measures the steel itself—strain, vibration, and deflection—rather than just temperature and humidity. Our guide to continuous SHM sensor monitoring explains how corrosion sensors, strain gauges, and modal analysis are combined into a single alarm system that flags coating degradation and section loss between manual inspection cycles.
| Corrosion Rate (mpy) | Corrosion Rate (mm/yr) | Severity | Recommended Action |
|---|---|---|---|
| < 5 mpy | < 0.13 mm/yr | Low | Maintain current plan |
| 5 – 20 mpy | 0.13 – 0.5 mm/yr | Moderate | Shorten interval; local patch repair |
| 20 – 50 mpy | 0.5 – 1.3 mm/yr | High | Full remaining-life re-evaluation |
| > 50 mpy | > 1.3 mm/yr | Severe | Derate use or strengthen urgently |
Classification bands are planning guides; structural action thresholds depend on section adequacy. consult our engineers to convert your UT trend into a remaining-life number.
Turning UT thickness trends into a steel building remaining service life rating requires more than corrosion rate alone: fatigue cracking, section loss at critical joints, and live-load reductions all feed the same calculation. Our remaining service life guide walks through how to combine NDT data with structural capacity checks to decide whether to repair, upgrade, or plan for demolition.
Inspection Frequency & Reporting
How often to inspect tracks the aggressiveness of the environment. A dry, indoor building can go about 5 years between full NDT surveys. Outdoor or coastal C3 buildings are better at 2–3 years. Heavy industrial or coastal C4–C5 buildings should be surveyed every 12–18 months. Crane girders, long-span roof trusses, and bridge-like nodes get closer attention than run-of-mill columns. (This is the frequency question; the full maintenance calendar is the separate topic in steel structure corrosion maintenance schedule.)
A useful inspection report is a document an engineer and an insurer can both act on. It should include: a location map of every test point (grid line and elevation); the UT thickness data plus the year-over-year corrosion trend; MT/PT defect records with location, size and severity grade; a DFT spot map with the out-of-spec points flagged; and a clear recommendation—repair scope, next inspection date, and whether a structural re-analysis is needed. The same reporting discipline applies at handover—see steel building site acceptance inspection—and when vetting a fabricator, see steel factory audit checklist and the practical steel building inspection checklist.
| ISO Category | Typical Environment | Full NDT Frequency | NDT Methods Required |
|---|---|---|---|
| C1–C2 | Dry indoor / inland | ~5 years | Visual + spot UT + DFT |
| C3 | Urban / light industrial | 2–3 years | UT grid + MT + DFT |
| C4 | Coastal / chemical | 12–18 months | UT grid + MT + DFT + coupon/ER |
| C5 | Heavy splash / marine | 12 months or less | UT + MT + PT + continuous monitoring |
Suggested intervals; tighten them after an over-limit reading, flood, or coating failure. Verify against your exposure and coating specification.
Conclusion
A steel structure corrosion inspection is UT for wall loss, MT for surface cracks, DFT for coating thickness, and mpy math for the rate of deterioration. A visual walk-around only finds what has already surfaced; NDT is what quantifies the internal loss. Remember that intact-looking paint does not prove structural safety—hidden corrosion is settled by UT numbers—and that the corrosion-rate trend across surveys matters more than any single absolute reading. Treat a recurring steel structure corrosion inspection as a data-driven event, then convert the trend into remaining life before you schedule the next repaint.
Know How Much Steel Is Left, Not Just How It Looks.
We arrange objective NDT—UT wall-thickness grids, MT on critical welds, DFT spot checks—with a written report that your engineer and insurer can both act on.
🏭 Explore: Steel Workshop · Steel Factory
Case Example
A coastal warehouse in Southeast Asia, 11,000 m2 (about 118,000 ft2) and 18 years old in a C4 environment, showed visible rust at the column bases and vague paint complaints. Rather than a full repaint, the owner commissioned a data-based inspection: ultrasonic thickness surveys at 40 points, magnetic-particle crack checks, dry-film coating thickness measurements, and comparison to a three-year-old baseline. The corrosion rate measured 0.12 mm per year (4.7 mpy) in the wet zones, below 0.05 mm per year (2 mpy) in the dry zones, and two base plates had lost 1.6 mm (1/16 in) of wall. A localized blast-and-recoat instead of full repaint saved about 45,000 USD, and the re-inspection was set at 24 months. Measurement, not appearance, drives the repair; see corrosion protection and coating inspection and testing for the NDT and DFT methods behind the decision.
Reference Links
- AISC 360 Specification for Structural Steel Buildings
- ASCE 7 Minimum Design Loads and Associated Criteria for Buildings and Other Structures
- ISO 12944 Corrosion protection of steel structures by protective paint systems
About the Author
Senior Structural Engineer
With over 20 years of hands-on experience in steel structure design and prefabricated building engineering, our in-house senior structural engineer has personally contributed to more than 500 steel building projects—including warehouses, industrial factories, aircraft hangars, agricultural buildings, and commercial structures. The focus is on translating design codes such as AISC 360, ASCE 7, and Eurocode 3 into buildable, cost-effective steel solutions that balance structural performance, fabrication efficiency, and total project cost.
Learn more about our engineering team
Frequently Asked Questions
What is the difference between corrosion inspection and corrosion protection?
Corrosion protection (covered in our protection article) is choosing the right coating, galvanizing, or cathodic system at design time. Corrosion inspection measures—after the building is up—how much steel has actually been lost, where cracks are forming, and whether the coating is still doing its job. One is preventive; the other is diagnostic.
Which NDT method should I use to check for rust under paint?
For internal wall loss under intact coating, use ultrasonic testing (UT)—it penetrates through paint and measures remaining steel thickness from one side. For surface cracks, use magnetic particle testing (MT) on carbon-steel welds. Dye penetrant (PT) is a cheaper fallback for non-ferrous or quick surface screening.
What is a safe corrosion rate for a steel building?
Below 5 mpy (0.13 mm/year) is generally low. 5–20 mpy (0.13–0.5 mm/yr) is moderate—shorten inspection intervals. Above 20 mpy (> 0.5 mm/yr) is aggressive and warrants a full remaining-life evaluation. Always compare readings year-over-year; the trend matters more than any single number.
How thick should my coating dry film be?
Typical epoxy systems run 120–250 μm (4.7–9.8 mils) total DFT, depending on the ISO 12944 exposure category. As a rule, at least 80% of spot readings must meet the design value, and no reading should exceed 200%—over-thick coating cracks and delaminates just like under-thick coating.
How often should I inspect a coastal steel building?
In coastal (ISO C4–C5) environments, plan a full NDT survey every 12–18 months—UT on column bases and truss nodes, MT on critical welds, DFT spot checks. Dry inland buildings can stretch to 5 years. Heavy industrial or splash-zone locations need even more frequent monitoring.
steel-structure-load-combination
steel-structure-fatigue-assessment