Engineering & Technology · Civil Structures & Structural Engineering

Construction stages, connections and inspection evidence

The completed structural system may not exist while it is being assembled. A floor can be present before its connections, restraint or material properties match the final model. Construction therefore changes the system being analysed, rather than merely moving it closer to a finished photograph.

Use Statics & Solid Mechanics, Materials Selection & Degradation and Measurement, Units & Engineering Evidence. Floor pressures, member properties, oscillator inputs and remaining areas are hypothetical teaching assumptions. No site-specific hazards, current code values, construction permissions or qualified structural designs are supplied.

Back to the chapter

After reading, you can

Explain how configuration, material development, connections and traceable inspection evidence affect structural behaviour during construction.

Analyse the configuration that actually exists

Identify which members and restraints are engaged at each relevant stage and which loads they receive. Temporary supports have finite stiffness, and their foundations also deform. In multistory concrete work, fresh material, developing slab properties and interacting supports can redistribute load between levels. The final arrangement cannot by itself establish a valid intermediate arrangement.

An early-age concrete stiffness and a strength assessment answer different questions. A sufficiently stiff path can attract force without having sufficient resistance; strength gain also depends on material and conditions. Original spring models can explain compatibility, but they do not specify a construction programme, support-removal time or acceptable temporary loading.

Follow force through the joint

A connection transfers the actions assumed in the load path. Bolts, welds, bearing surfaces, reinforcement anchorage and bonded interfaces provide different transfer mechanisms. Joint eccentricity can introduce a moment; deformation or slip can change the boundary condition seen by adjacent members. A symbol for a rigid or pinned joint is a modelling decision that needs physical evidence.

A member’s visible presence does not prove its connection is complete or that a restraint is engaged. Assembly tolerances and access also matter: a detail hidden after finishing may require evidence before concealment. Inspect the actual configuration against the relevant drawing revision and acceptance basis rather than infer completion from the intended sequence.

In an abstract one-coordinate model, path A has stiffness 10 MN/m and path B 20 MN/m. The same zero-displacement reference and no initial load or gaps are assumed for both configurations. A alone gives δ=P/10 mm for P in kN. A+B with compatible displacement gives δ=P/30 mm. At 60 kN, A alone displaces 6 mm and carries 60 kN; both together displace 2 mm with A carrying 20 kN and B 40 kN. Curves compare model configurations, not a history that engages an unstressed support after loading, and supply no shore-removal permission.

Keep observations separate from acceptance conclusions

Inspection records should identify the component and location, condition, date, method and controlling document. Photographs can document visible geometry but cannot establish every hidden material or joint property. Dimensional checks, material records and suitable nondestructive examinations answer different questions. Record limitations and unresolved deviations so later reviewers can understand what was actually verified.

A measured value needs a stated criterion and appropriate uncertainty before it supports an acceptance decision. A crack-detection method and a method for observing crack activity need not provide the same evidence. Historical FHWA evaluations demonstrate that equipment can have detection and interpretation limits; they do not establish current equipment capability or a universal probability of detection. Unexpected findings need an assessment of their significance for the affected load path.

Two available paths change the force distribution

Compare two abstract unloaded configurations under a common applied 60 kN force. Path A has linear stiffness 10 MN/m; path B has 20 MN/m. Assume a shared zero-displacement reference, no gaps or initial forces and a common displacement when both are present. A alone gives δ=60000/10000000=6 mm and carries the entire 60 kN.

With A and B together, combined stiffness is 30 MN/m and displacement is 2 mm. A then carries 20 kN and B 40 kN. The curves show displacement against load for these separate configurations. They do not model adding an initially unstressed support after the system has already deformed; that history requires different compatibility and initial-state information.

Try the reasoning

Can a final-state analysis alone establish the adequacy of an earlier unbraced configuration?
Show the answer and reasoning
No. The restraints and possibly the actions differ. The earlier state needs its own model and evidence.
If B is twice as stiff as A in the stated parallel model, do both carry 30 kN?
Show the answer and reasoning
No. Compatible displacement gives a 1: 2 force split: 20 and 40 kN under the 60 kN total.
Can a photograph of the outside of a joint prove that an inaccessible weld has no internal defects?
Show the answer and reasoning
No. It records the visible condition. Internal-defect claims require an appropriate examination and its stated limitations.

What to carry forward

Treat each stage as a real structural configuration and preserve evidence for the connections and properties its model assumes.

References

Teaching resources and research supporting this lesson’s explanations and examples.