Engineering guide

Tank Settlement and Bulging Assessment to API 653 and API 579

A tank settlement and bulging assessment determines whether a storage tank's foundation movement or shell distortion is still within acceptable limits, or whether repair, re-levelling or a reduced fill height is needed. This guide covers the survey data required, how API 653 Annex B classifies settlement, how shell bulges are assessed with API 579, and where finite-element analysis fits.

Last updated:

Why tanks settle and shells bulge

Every aboveground tank settles to some degree, because the hydrostatic load of the product compresses the soil beneath it. Settlement becomes a problem when it is uneven: around the shell circumference, between the shell and the centre of the bottom, or locally under part of the floor.

Shell bulges and dents have several causes. Differential settlement distorts the lower courses; overfilling or blocked vents can apply pressure or vacuum the shell was not designed for; and fire, impact or poor repair welding leave local deformation. The assessment route depends on which mechanism is active, so the cause should be identified before acceptance limits are applied.

Soil conditions matter too. Tanks on soft clays, reclaimed coastal ground or poorly compacted ringwall fill can keep settling for years, while tanks on rock or well-prepared granular pads usually stabilise early in their life.

Survey data you need

Any assessment is only as good as the measurements behind it. API 653 sets out how elevations should be taken around the shell, with a minimum number of points and a maximum spacing between them, and recommends internal bottom surveys when the tank is out of service.

Laser scanning is now common for shell distortion because it captures the full surface rather than a few vertical lines. It is especially useful for bulges, where the shape of the deformation matters as much as its peak value.

Typical data for a settlement and bulging assessment
DataPurpose
Shell elevation survey at equally spaced pointsUniform, tilt and differential settlement of the shell
Internal bottom elevation surveyEdge settlement and bottom depressions or bulges
Shell verticality and roundness readingsOverall distortion and effect on floating-roof seals
Laser scan or detailed profile of local bulgesGeometry for API 579 or FEA assessment
Shell thickness surveyActual thickness for stress calculations
Original design data and drawingsCourse thicknesses, materials and design fill height
Previous surveysRate of settlement and whether it is still active

Settlement types in API 653 Annex B

Annex B of API 653 separates settlement into components, because each has a different effect on the tank. The measured shell elevations are first fitted with a best-fit plane, and the deviation from that plane is then evaluated.

Uniform settlement is the whole tank moving down evenly. It does not stress the shell, but it can strain connected piping and affect drainage. Rigid-body tilt, or planar tilt, rotates the tank as a whole; it raises the liquid level on one side and can affect floating-roof operation and nozzle loads. Out-of-plane, or differential, shell settlement is the deviation remaining after tilt is removed, and this is what distorts the shell. Annex B gives an allowable out-of-plane deflection that depends on the spacing of the survey points, the shell material's yield strength, its elastic modulus and the tank height.

Bottom settlement is assessed separately. Edge settlement is the floor dropping relative to the shell near the corner joint, and Annex B provides limits that depend on the width of the settled zone and the orientation of bottom lap welds. Local depressions or bulges in the floor away from the shell are assessed against a limit tied to the size of the affected area.

Assessing shell bulges with API 579

API 653 gives tolerances for shell shape, but it does not provide a full method for assessing an existing bulge or dent. For that, engineers normally use API 579-1/ASME FFS-1, which contains a part dedicated to weld misalignment and shell distortions, including out-of-roundness and general bulging.

The Level 1 and Level 2 methods in that part use measured geometry to estimate the additional bending stress caused by the distortion and compare it with allowable values. Where the distortion is irregular, combined with local thinning, or close to a nozzle or the bottom corner, a Level 3 assessment with finite-element analysis is normally used. Our guide to API 579 assessment levels explains how the three levels differ.

Where finite-element analysis is used

FEA is the right tool when the simpler methods do not fit the geometry or are too conservative to give a workable answer. A typical model builds the shell from the laser-scan surface, applies hydrostatic pressure for the design liquid and fill height, and includes the bottom and the relevant course thicknesses.

Results are evaluated against the stress categories and limits agreed in the assessment basis, and the model can be run at several fill heights to find a safe operating limit. In one of our case studies, a static FEA of an existing tank shell under a high-SG product established a clear maximum filling height for the owner. Our tank FEA verification case study shows how shell and roof models are built and checked, and our project references include tank bulging assessments to API 650 and API 653 and forensic bulging investigations.

Two modelling choices deserve care. Boundary conditions at the bottom-to-shell junction strongly affect stresses in the lowest course, so the bottom plate and its contact with the foundation should be represented rather than replaced by a fixed edge. And where the bulge is large, a geometrically non-linear analysis may be needed, because the distorted shell tends to straighten under pressure and a linear model can overstate or understate local bending.

Common mistakes

Each of these leads either to an unsafe acceptance or to an unnecessary repair. A short review of the survey data before the assessment starts avoids most of them.

  • Too few survey points, so a localised dip between readings is missed
  • Comparing raw elevations with the limit instead of the deviation from the best-fit plane
  • Treating one survey as proof of stability when settlement may still be active
  • Assessing a bulge without first establishing whether it came from settlement, pressure, vacuum or fire
  • Ignoring the effect of tilt on floating-roof seals, rolling ladders and nozzle piping
  • Using nominal thickness where the thickness survey shows corrosion

Outcomes and when to involve an engineer

The result of an assessment is usually one of four outcomes: acceptable as is with a re-inspection interval, acceptable at a reduced fill height, acceptable after local repair such as re-levelling or plate replacement, or not acceptable without major repair. API 653 requires settlement beyond the Annex B limits to be evaluated by an engineer experienced in tank settlement, and any repair must follow the owner's inspection programme.

SLETEC carries out settlement and bulging assessments from survey data supplied by the owner's inspector or survey contractor, and prepares repair packages through our tank repair engineering service when repair is needed.

Frequently asked questions

What is the difference between tilt and differential settlement?

Tilt is a rigid-body rotation of the whole tank and does not by itself distort the shell. Differential, or out-of-plane, settlement is what remains after the best-fit plane is removed, and it is this component that bends the shell and is checked against the API 653 Annex B limit.

Can a bulged tank shell stay in service?

Often it can, if an API 579 assessment or FEA shows that the stresses from the distortion are acceptable at the operating fill height. The assessment may set a reduced fill height or a re-inspection interval. The cause of the bulge must also be understood and stopped.

How often should tank settlement be surveyed?

The interval is set by the owner's inspection programme under API 653 and depends on the tank's history. Newly built tanks and tanks with known settlement are usually surveyed more often. Comparing several surveys shows whether settlement is still active.

Is laser scanning necessary for a bulging assessment?

It is not mandatory, but it is the most practical way to capture the full shape of an irregular bulge. Manual profiles can be enough for small, simple distortions. For an FEA assessment, scan data gives a much more reliable model.

Discuss your project with an engineer

Send drawings, datasheets or a short scope – an engineer replies with scope, schedule and price, usually within 24 hours.