Structural Steel Design / A complete steel building
What governs where
Eighteen modules have each taught a check. This lesson puts four members side by side and asks which check actually decides each one — and the answers are all different.
Lesson review: expert verification remains open
This lesson has received a full-context editorial review and a check of its technical explanations. Selected arithmetic has been checked separately. This is not independent engineering sign-off.
Corrections checked in this pass
- Historical SC/PC escalation and fixed NDT/cost claims removed from lessons; public execution-class calculator already withheld. Current first-generation selection basis and EN1090-2 amendment identified; traceability distinguished from sign-off.
Checks needed for project use
- A steel-design reviewer must check the first-generation calculation profiles, second-order analysis assumptions, connection rules and execution references against the applicable standards and amendments.
- Project execution specification, applicable edition/National Annex, weld-specific inspection/tolerance requirements and actual fabrication controls. Legacy executionClass function remains in library but is not imported by the public lab; do not restore it as a current recommendation engine. Full independent engineering sign-off remains open.
These are educational models and exercises. Check their assumptions before using the results. The full technical content and every diagram have not yet been independently verified.
Concrete and steel calculations retain their stated first-generation Eurocode teaching basis. Use the standards and amendments specified for your project; do not mix generations without an appropriate design basis.
Read the lesson register and outstanding review notesWhy this matters
A course is made of modules, and a module teaches one thing at a time. That is the only way to learn the material, and it has one serious side effect: it makes every check look equally likely to matter.
They are not. In a real building, each member type has a limit state that decides it, and those limit states are different for different members. A designer who has absorbed the modules but not the pattern will check everything for bending, size everything for strength, and produce a building that is wrong in several different directions at the same time.
This lesson puts four members from one building next to each other and asks a single question of each: what actually decides this?
By the end of this lesson you should be able to
- Name the governing check for each member type and say why
- Read the gap between the top two checks, and act on it
- Recognise a member carrying steel that is not working
- Say why each governing limit state has a different remedy
What you should already know
- Deflection and serviceability (Module 5)
- Bending, shear and lateral-torsional buckling (Modules 6 and 8)
- Flexural buckling of columns (Module 7)
- Fatigue (Module 17)
Four members, one building
The building is a three-storey industrial and office structure: 9 m floor beams on a braced frame, with a crane bay along one side. Nothing exotic; the sort of thing that gets built constantly.
Take one member of each kind and assemble every check that applies.
The floor beam. Bending at 39%, shear at 14%, lateral-torsional buckling at 40%, deflection at 56%. Deflection governs, and by a clear margin. Nothing here is close to failing — but if the span grows, or the loading rises, it is deflection that arrives first.
The column. Flexural buckling at 88%, bending at 13%. The column is a stability problem, and it is the tightest member in the building.
The brace. Flexural buckling at 33%. It is nowhere near working, which is normal: braces are usually sized by a maximum permitted slenderness, by the connection, or simply by what is available, rather than by the force in them.
The crane girder. Bending at 54%, fatigue at 335%. It fails, and it fails on a check that has nothing to do with strength.
Three different limit states govern across four members. Each one has a different remedy, and applying the wrong remedy is worse than useless — it costs money and moves nothing.
Notice also that two of the four members are running below 60%: the brace at 33% and the floor beam at 56%. A low utilisation is information rather than a verdict. Neither of those members is sized by the force in it — one by slenderness and availability, the other by a deflection limit — and both are perfectly normal. What a low utilisation asks for is a reason, not a smaller section.
Worked example
The same beam, decided by three different checks
Given
- A 610×229×113 UB floor beam carrying 30 kN/m
- Deflection limited to span/360
- Restraint spacing and span both variable
Find
Which check governs, and what changes it
Try it
What governs where
Four members in one building. Each bar is a member's governing check, and the label on it is which limit state that is. They are not the same limit state — which is the whole argument of this module.
Inspect a member
- Distinct governing limit states
- 3
- Which ones
- Deflection, Fatigue, Flexural buckling
- All members pass
- NO
- Failing
- crane-girder
- Carrying steel that is not working
- floor-beam, brace
- — inspecting —
- Floor beam
- Deflection
- 56 %
- Lateral-torsional buckling
- 40 %
- Bending
- 39 %
- Shear
- 14 %
- Gap to the next check
- 16 points
- Top two are close
- no
Deflection governs clearly. Deflection is a stiffness problem, and stiffness comes from depth. A stronger steel does nothing: E is the same for every grade. Go deeper, or accept a longer span with a different floor system.
Things worth trying
- Start at the defaults — 9 m span, 3 m restraint spacing, category 71 weld. Read the top chart first: three different limit states are governing across four members.
- That is the argument of the module in one picture. A designer who checks everything for bending has made four different mistakes, not one.
- Inspect the floor beam. At 9 m it is not a bending problem at all — deflection governs, and bending is well behind. Read the advice: stiffness comes from depth, and a stronger steel does nothing because E is the same for every grade.
- Now walk the span out to 15 m. Deflection grows with the fourth power of span while bending grows with the square, so the gap widens rather than closing.
- Bring the span back to 9 m and take the restraint spacing out to 9 m. LTB takes over from deflection — a completely different remedy, reached without touching the section.
- Find a restraint spacing where LTB and deflection are within ten points of each other. The verdict changes: this is now a section problem, because relieving one hands the governing role straight to the other.
- Inspect the brace. It runs at about a third, and the audit flags it as carrying steel that is not working — braces are usually sized by minimum slenderness or by what is available, not by force.
- Note that the FLOOR BEAM is flagged too, at 56%. A low utilisation is information, not a verdict: a beam governed by deflection is often under 60% on every strength check, and that is not waste — it is what a serviceability-governed member looks like.
- Inspect the crane girder at category 71. Fatigue is at 335% while bending sits at 54% — it fails by a factor of three on a check that is not bending at all.
- Now raise the weld detail category to 125. Fatigue falls to 61% and the girder passes. The section never changed; only the quality of one weld detail did.
Practice
A floor beam has bending at 82%, shear at 31%, LTB at 79% and deflection at 44%. What is the gap between the governing check and the next tightest, in percentage points?
Check yourself
A brace in the stability system is running at 33% of its buckling resistance. What does this most likely indicate?
Summary
- Four members in one building were governed by three different limit states
- The 9 m floor beam is a deflection problem at 56%, with bending at 39%
- Taking the restraints to 9 m hands the governing role to LTB at 89% — the beam never changed
- Near 6 m spacing the two are within a point: a section problem, not a single-check problem
- Read the GAP as well as the governor — it decides whether there is one thing to fix
- Four of the five common remedies do not match the instinct to use a bigger, stronger section