A tank floor corrodes from the side you cannot see, sitting on damp ground, and it gives no warning at all. The internal inspection that finds it is the single most expensive item in the API 653 cycle, and parts of the floor still get missed.
A storage tank floor is a large area of relatively thin plate with product on one side and damp soil on the other. The soil side is where most of the damage happens, and there is no way to look at it without emptying, cleaning and entering the tank.
That is why an API 653 internal inspection is scheduled in years and costed in hundreds of thousands. Everything below happens inside that window.
A scanner is walked across the floor plates. Powerful magnets saturate the plate; where metal is missing from either surface, flux leaks out and sensors detect it. A trained operator covers a large floor in a shift or two.
What it does well: finds underside corrosion across the whole scanned area. This is the only practical way to survey a floor, and it turns an impossible problem into a manageable one.
What it does not do: measure. MFL reports an indication with an estimated severity band, typically as a percentage of plate thickness. It cannot distinguish top-side from bottom-side loss without a separate top-surface sensor, and its sizing carries real tolerance.
Which is why every significant MFL indication gets UT prove-up: a manual ultrasonic measurement at the indication to establish an actual remaining thickness. The MFL finds it; the UT measures it. Neither replaces the other.
This is the section that matters, because these areas are also where corrosion concentrates.
| Area | Why MFL struggles | What to do instead |
|---|---|---|
| Critical zone — the annular ring at the shell | Scanner cannot get close to the shell-to-floor weld; geometry and stiffness change | Manual UT, and often removal of a coupon. This is the highest-consequence area on the floor. |
| Lap welds | The step in the plate lifts the scanner and disturbs the flux path | Manual UT along the lap; vacuum box for through-leaks |
| Patch plates from earlier repairs | Two thicknesses of steel; MFL reads the combination | Manual UT, and treat the original plate underneath as unknown |
| Under sump and column bases | Physically inaccessible | Visual and manual UT where reachable; note the gap in the report |
| Heavily pitted or scaled surface | Scanner cannot maintain contact and coupling | Grind and re-scan, or switch to manual mapping |
The annular ring near the shell carries the highest stress in the floor, because that is where the shell load transfers in. It is also where water tends to collect under the tank. And it is the area MFL covers worst. An inspection report that gives a floor a clean bill from MFL alone has not inspected the part most likely to fail.
A transparent box with a soapy film over a weld, evacuated. Bubbles mean a through-leak.
This is not a thickness measurement and it is not a corrosion survey. It answers one question: is this weld leaking right now? Applied to floor welds after repair and to areas of concern, it is the direct check that the floor is tight.
API 653 sets minimum remaining thickness for floor plate and a separate, stricter treatment for the critical zone. The inspection produces:
Soil-side attack depends on what is under the plate: moisture, soil resistivity, whether the cathodic protection is working, whether the release-prevention barrier has failed, and whether there is a puddle under one part of the tank. A rate averaged across the floor can be meaningless. The rate that sets your interval should come from the worst area, not the average one.
Inspection finds soil-side corrosion; it does not slow it. The things that do are all under the tank and all decided at construction or at the last floor replacement: a competent release-prevention barrier, a properly designed and maintained cathodic protection system, a foundation that drains rather than holds water, and a chime detail that does not channel rain underneath.
Of these, cathodic protection is the one that quietly stops working. A CP system installed twenty years ago and never surveyed is a system nobody can claim is protecting anything — and its failure is invisible until the next internal inspection, which may be a decade away.
What underside corrosion does to a floor plate, what causes it, and every method in the guide that detects it.
Open the mechanism libraryWho writes this. A mechanical engineer with twelve years in oil and gas — in-line inspection, fired heater and furnace inspection, and pipeline integrity. What is here comes from the published standards and from what those years in the field actually looked like. It is not written by an API-certified inspector.
This is not an assessment. Nothing on this site may be used to justify a decision about real equipment. Assessing plant requires the current editions of the applicable codes, data from a licensed source, and a competent engineer who signs for the answer. · Integrity Field Guide