How to design an RC column with and without second-order effects (EN 1992-1-1:2023)

A reinforced concrete column is designed for the axial force \(N_{Ed}\) and a design moment \(M_{Ed}\). Whether \(M_{Ed}\) is just the first-order moment or also contains a second-order part decides how much steel the column needs. EN 1992-1-1:2023 gives two criteria for that decision. Formula (7.25) in 7.4.1(3) looks at the whole frame: the buckling load of the structure must be at least ten times the design vertical load. Formula (O.12) in Annex O looks at one column at a time: its slenderness must not exceed a limit. This article first sets out the design method step by step, with and without second-order effects, then works it through on one one-storey, one-bay frame, and finally compares where the two criteria let second-order effects be neglected and what design moment follows. In the (7.25) route the frame criterion only fixes the effective length, (O.9) braced or (O.10) unbraced, and every column is then checked locally with (O.12) using that length.

Short answer. With fixed bases, (7.25) allowed second-order effects to be neglected up to a column force about three times larger than (O.12) used as the decision rule. In that range the (7.25) route gave the smallest design moments: with the braced length the local (O.12) check is satisfied with a wide margin, while the (O.12) route added 14–38 % (O.7) or 4–14 % (O.8). Once (7.25) says sway, both routes use the unbraced length and the same local check, so they give the same design moment. Neither of them sees the frame approach instability: near \(\lambda_{cr} = 1\) the nominal-curvature moment (O.7) fell well below a global magnification of the frame, the moment-magnification method (O.8) did not.

Updated 15 November 2026: the (7.25) route now follows the FaxinaX column check — (7.25) decides the effective length and the local check (O.12) of each column decides on the second-order moment. The earlier version magnified the moments of a sway frame globally with (O.23); that variant is kept for comparison.

The buckling comparison also includes a four-storey, three-bay RC frame with an effective global creep coefficient of 2 and an illustrative Annex O.3 calculation, with its applicability limits explained.

Subscribers only

The introduction is free. The full article and the book chapters are for subscribers, from $1 per month.

Subscribe for $1 / month Sign in