Home / Services / Steel structure design (structural and detail drawings) / Corrosion wear and the load-bearing capacity of steel members: how section loss is accounted for
Steel structure design (structural and detail drawings)
Corrosion wear and the load-bearing capacity of steel members: how section loss is accounted for
Rust does not just spoil the look: it reduces the section, and with it the load-bearing capacity and stiffness of the member. Below are the kinds of corrosion damage, which of them the calculation takes into account and which it rules out, how section loss goes into the check, and what signs lead to a member being strengthened or replaced. There are no formulas here on purpose: they depend on the section shape and belong to the designer.
Contents
What corrosion does to a member
Corrosion acts in two ways. The first is obvious: the section gets thinner, and the stress in the remaining metal rises under the same load. The second is less visible: in a statically indeterminate structure the members lose stiffness unevenly, and the forces move to those that have kept better. So checking one corroded member on its own may underestimate the load on its neighbor.
A thin section loses a bigger share of its area to the same thinning than a thick one. That is why angles, channels and thin plates in damp places fail earlier than massive sections.
A layer of rust has no load-bearing capacity. Only the metal that is left counts in the calculation, measured after cleaning down to a bare surface.
Types of corrosion damage
Electrochemical corrosion with the help of moisture prevails in industrial buildings. It is divided into general corrosion, uniform or uneven over the surface, and local corrosion: patches, ulcers, pinpoint attack, through-penetrations. Local corrosion appears where the action is local: a leaking roof, a leaky pipe, a member that is constantly wetted where it touches damp masonry or concrete.
| Type | What it looks like | How it is treated in the assessment |
|---|---|---|
| General, uniform | A layer of rust and thinning roughly equal over the whole surface | The thinning is measured and entered into the calculation |
| General, uneven | Thinning varies, with zones of deeper attack | Measured at several points and entered into the calculation taking the unevenness into account |
| Local (patches) | Separate areas of heavy attack, the rest of the surface clean | Checked by calculation as a local weakening |
| Ulcer-type | Separate pits a few millimetres across | Guideline for the serviceable state: ulcer depth up to 0.5 mm, diameter up to 2 mm |
| Pinpoint (pitting) | Small deep points | Guideline for the serviceable state: diameter up to 0.1 mm; depth is taken into account by a pitting factor |
| Intergranular | Destruction along the grain boundaries, may be almost invisible from outside | Not permitted |
| Subsurface | Attack under the outer layer of the metal | Not permitted |
| Stress corrosion cracking | Cracks under the combined action of stress and environment | Not permitted |
| Crevice | In gaps and joints where moisture gets in | Category B defect: no direct danger, but it develops; crevices are sealed |
How section loss goes into the calculation
First the remaining thickness of the metal is measured, not that of the rust layer: after cleaning down to bare metal, at several points along the length and around the perimeter, recording the worst zones. The difference between the original thickness (from the drawing or the rolling table) and the measured one is the thinning.
Then the geometric properties of the section are reduced: area, moment of inertia, section modulus. Every surface in contact with the environment is reduced by the depth of corrosion penetration. For a closed section corroded on one side this depth equals the thinning. For an open section (I-beam, channel, angle) that corrodes on both sides, half of the measured thinning is taken for each surface.
How many per cent of the section this takes depends on how much metal there is per unit of perimeter in contact with the environment. The thinner the wall and the more open surfaces, the faster the loss grows. So the same depth of corrosion hits an angle, a channel, an I-beam and a tube differently. The designer applies the formulas for each shape under the current methodology for verification calculations.
Pitting is taken separately. The pitting factor depends on the depth of the attack and on how densely it is spread. The steel for calculating corroded members is characterized by tests, not only by the certificate.
Then both groups of limit states are checked: strength and stability under the design loads, and also deflections and movements, because stiffness has fallen too. In statically indeterminate systems the whole scheme is recalculated, not one member, since a weakened member passes part of its force to its neighbors. Through attack and cut-outs are calculated as a reduction of the net area, but through corrosion is not entered into the calculation as acceptable: it is removed.
If the conditions of strength, stability or stiffness are not met, the weakening is treated as corrosion damage and the structure is strengthened. The strengthening is then calculated on the remaining section. The general logic of a verification calculation is covered in the article Condition assessment of steel structures.
When a member is strengthened and when it is replaced
There is no single limit for all members. The decision depends on what kind of member it is, how much section is lost, where exactly (in the most stressed zone or not), the state of the rest of the member, whether the zone can be reached, and what the recalculation with strengthening shows.
There is a guideline from the repair of crane steelwork: a member that has lost more than 5 % of its section area to corrosion in rod systems, or more than 10 % in beam systems, is treated as unfit for repair and replaced. This is a narrow criterion for lifting cranes. For building structures and poles it is carried over only with the justification of the project designer.
Strengthening is considered when the loss is local, the rest of the member is sound, the zone is accessible and the recalculation with strengthening works out. Columns are repaired in the case of substantial corrosion wear or significant local damage, and this is usually done under load because they are hard to unload. Replacement is used when strengthening is impossible, and the earlier structural solution is usually repeated.
The strengthening metal must not be inferior to the existing one in mechanical properties, toughness and weldability, and in an aggressive environment its corrosion resistance is not lower than that of the existing metal. If the structure was never welded before and there are no weldability data, weldability is assessed first (see clarifying steel properties).
Cleaning off rust can itself weaken a badly corroded member for good. So temporary props are placed under such members for the time of cleaning. For metal poles similar decisions are covered in the article Wear and Corrosion of Metal Poles.
Strengthening must not create new corrosion sites
Plates and cover plates in corroded zones easily form crevices and closed cavities where moisture stays. So the strengthening is designed so that every surface can be reached for protection, and crevices are sealed. At the cut-off ends of strengthening members sharp stress concentrations are avoided, and welds are not bunched together.
Welds on strengthening members are made continuous. Intermittent (stitch) welds are acceptable only in statically loaded structures in a non-aggressive environment, and under dynamic loads they are not used.
If the damage was caused not by moisture but by vehicle impacts or heat radiation from equipment, the strengthening design provides protective casings and guards, otherwise the damage will repeat.
Protection after repair: paint and zinc
Without protection the strengthening is pointless: the mechanism that ate the metal will start working again. The protection system is chosen by the aggressiveness of the environment and the access to the surfaces. A paint coating remains the shortest link in the protection of a frame: it is renewed much more often than the metal itself is replaced.
For new members made in the shop one option is hot-dip galvanising. How zinc works and how long the coating lasts is covered in the articles Hot-dip Galvanising Service Life and Designing Steel Structures for Hot-dip Galvanising. A member is galvanised whole in the bath, so for a structure that is already erected this means dismantling. On site it is protected by a paint system, and the weld areas of strengthening parts welded on site are protected separately.
What we can do
Measuring the remaining thickness and assessing the members is done by technical inspection. POLYGONAL prepares KM and KMD drawings for strengthening or replacement (design), fabricates members to the customer’s drawings (fabrication), and carries out hot-dip galvanising and powder coating. Erection on site is done by our partner construction company.
Standards and validity
The assessment of corrosion wear is part of technical inspection under the current procedure and the Methodology for conducting an inspection and recording its results (approved by the order of 06.08.2022 No. 144) and DSTU 9273:2024. Corroded members are calculated under DBN V.2.6-198:2014 «Steel structures. Design standards» and DBN V.1.2-14:2018 «General principles of reliability and structural safety of buildings and structures» (both with Amendment No. 1).
The classification of environmental aggressiveness and the requirements for corrosion protection systems are set by separate current standards: the designer confirms their titles and editions before the project. The guidelines given in the article (sizes of ulcers and pits, the percentage of section loss for crane structures) are for reference and do not replace the decision of the project designer. The validity and current edition of all the documents mentioned need to be checked.
Updated:
Frequently asked questions about corrosion wear of steel members
How many per cent of corrosion is considered dangerous?
There is no single figure for all structures. The loss becomes dangerous when the calculation with the remaining section fails or when the attack is of a kind that is not permitted. For crane structures the guideline is 5 % of the section area in rod systems and 10 % in beam systems, but this cannot be carried over to other structures without justification.
Can it simply be cleaned and painted?
Only if the calculation with the measured remaining section works out. First the thickness is measured after cleaning, then the decision is made. Cleaning itself can weaken a badly corroded member for good, so temporary props are placed under it for the time of the work.
Does the layer of rust count as part of the load-bearing section?
No. Corrosion products have no load-bearing capacity. Only the metal that is left counts, measured after cleaning.
Related products: Steel structure design (structural and detail drawings) · Custom steel fabrication · Technical inspection
More on this topic: KM Drawings: Contents, Stages, Calculations · What is KMD: meaning and how it differs from KM · KMD Drawings: Contents, Benefits, Cost · KM and KMD Stages: Contents of the Sets · KM and KMD Drawing Rules: DSTU, Specification, Scales · Steel Structure Calculation: Methods and Checks
Get a quote for your requirements
Send us your requirements — we will calculate the price and lead time.