Elsevier

Engineering Structures

Volume 26, Issue 12, October 2004, Pages 1809-1819
Engineering Structures

Sway of semi-rigid steel frames: Part 1: Regular frames

https://doi.org/10.1016/j.engstruct.2004.06.018Get rights and content

Abstract

Lateral sway is most likely to control the design of semi-rigid steel frames where the frame arrangements do not include any form of bracing. This paper investigates the sway behaviour of semi-rigid regular steel frames, i.e. frames having the same arrangement of beam and column sections at all levels, and hence proposes some design charts for the prediction of sway that eliminate the need for doing any numerical modelling. Schueller’s equation has also been modified to incorporate connection flexibility in addition to its original rigid frame considerations. All the proposed methods have been validated using results obtained from numerical analysis.

Introduction

Limiting the lateral sway of building frames is an important design consideration, particularly in those cases for which so-called frame action is relied upon to provide the necessary lateral stiffness. For arrangements that do not include any form of bracing, it is the inherent bending stiffness of the beams and columns themselves that provides the necessary lateral stiffness. In recent years, there has been a growing realisation that a further structural component also contributes some degree of flexibility—both the connections between individual beams and columns and arrangements at the base of the columns where joints are made to the foundation system are now recognised as structural ‘members’ with their own strength and stiffness. Estimating the lateral sway of frame structures, including the making of realistic allowances for the contributions of the connections, is therefore an important design issue.

Techniques for analysing frames incorporating semi-rigid joint effects have been developed significantly during the last two decades. Much of this work has recently been reviewed [1], [2] and only a few particularly relevant contributions are covered herein. Huang and Morris [3] conducted a computer analysis on bare steel frames to study the effect of connection properties on lateral sway. The connection moment–rotation relationship was taken from test data. Colson and Bjorhovde [4] analysed a two storey, two-span bare steel frame assuming pin connections for the exterior beam-to-column connections and semi-rigid behaviour for the interior ones using the computer program PEP-Micro. Deirlein [5] studied semi-rigid bare steel frames using the computer program CU-STAND. All the beam-to-column connections were modelled using zero-length rotational springs to account for their non-linear moment–rotation behaviour. Li et al. [6] analysed a two-span, two storey semi-rigid bare steel frame using the general-purpose finite element software ABAQUS. They studied the effect of connection length and connection stiffness on the moments at different sections. Ye Mei-xin et al. [7] proposed a finite element model for composite frames using the general-purpose software ABAQUS.

It is, of course, possible to utilise a comparatively rigorous approach in which a computer-based frame analysis that explicitly allows for the contributions of the joints is undertaken. However, as with all computerised approaches, initial values must be assigned to the key member properties before such a study can be undertaken. It is, certainly, helpful if such values can be selected as being realistic in the sense that they are close to those likely to be the result of the final design decisions.

When checking deflections under serviceability conditions, it is important to recognise two features:

  • (i)

    Such calculations are normally undertaken assuming elastic behaviour.

  • (ii)

    The essential design requirement is that the calculated deflection be not greater than the permissible limit, i.e. great precision in undertaking the analysis is not necessary unless the calculations produce a result that is very close to the limiting value.

Methods that utilise the principles of elastic analysis in an approximate fashion are therefore likely to be useful when attempting to arrive at an appropriate overall balance between member sizes and joint configuration.

Ammerman and Leon [8], Ahmed [9] and Ahsan [10] proposed some methods for sway prediction of semi-rigid frames. The proposals made by Ammerman and Leon and Ahmed were obtained using some specific test data and were applicable only to low-rise frames. Ahsan’s [10] proposed method is applicable only to medium-rise frames ranging from five to eight storeys high. So a more general approach is warranted to predict the sway of semi-rigid frames.

This paper investigates the sway behaviour of semi-rigid steel frames using the finite element (FE) package ANSYS V5.4 [11] to generate results that are then used as the basis for a simplified method for sway prediction. The proposed method is based on design charts obtained from a thorough investigation of the key parameters that affect the sway response of semi-rigid frames. Modifications are also made to the approach, originally proposed by Schueller [12], for estimating the lateral sway of frames with rigid joints, that permits explicit allowance to be made for the influence of the flexibility of the beam-to-column connections. Further extension of this approach to deal with the influence of column base effects is in progress.

The accuracy of the proposed methods is demonstrated by comparisons against a portfolio of numerical results for a representative set of structures. These have been obtained using the ANSYS package.

Section snippets

Numerical modelling technique of semi-rigid frame behaviour and its verification

In the present study, a general-purpose finite element program ANSYS V5.4 has been used to model semi-rigid frames. The beams and columns were modelled using ‘BEAM3—2D Elastic Beam’ elements while a rotational spring element ‘COMBIN39—non-linear spring’ was used to model the beam-to-column connections. This spring has the capability to incorporate the non-linear moment–rotation behaviour of connections. The column bases were modelled as fixed.

The developed numerical models have been extensively

Objective and methodology of the present study

Proper understanding of the frame response to various factors is very important in order to make accurate predictions of sway. Frames having a wide range of variation in cross-sectional and geometrical properties were analysed to reveal the key parameters affecting the sway behaviour. The following sections describe the frames considered in the parametric studies reported in this paper.

Design charts for calculating flexibility factors

Sway behaviour of steel frames is largely affected by the Kc/Kb ratio (as reported by Ahsan [10]). The parametric studies carried out in the previous section revealed some other parameters—number of bays, number of storeys and the ratio of beam length to column height Lb/Lc—that affect the sway behaviour of comparatively taller frames, i.e. frames higher than 15 storeys. This leads to the following simplified design proposal.

Frames grouped into six categories according to the number of storeys,

Step 1: Determination of FF

Fig. 12, Fig. 13, Fig. 14, Fig. 15, Fig. 16, Fig. 17 give the FF for a particular semi-rigid frame. Their use is described below in a step-by-step manner.

  • (i)

    The number of storeys and number of bays for the given frame are obtained from the frame geometry.

  • (ii)

    By knowing the values of Lb/Lc and the number of storeys, the appropriate design chart(s) should be selected.

  • (iii)

    The Kc/Kb value for the given frame is determined using the following equationKcKb=IcLbIbIc

  • (iv)

    Knowing the value of Kb/Kj, the FF(s) can be

Sway prediction using modified Schueller’s equation

Schueller’s equation was originally proposed for the sway prediction of rigid frames considering the bending of beams and columns and axial deformation of columns. In the case of semi-rigid frames, an additional term is required to take into account the rotation φ of the connections. If Schueller’s equation is modified to incorporate connection rotations, the general form of the equation will beΔsemi-rigid=HVch212EIc+HVgL212EIg+2NcH23EAcB+MHKj

But under the working load, the moments and

Illustrations and comparisons

The design steps presented in Section 6 are illustrated here using worked examples. Two completely different frames were considered to cover all the relevant features of the methods, especially interpolation when using more than one design chart for sway prediction.

Conclusions

The key parameters affecting the sway behaviour of semi-rigid regular steel frames are identified and hence design charts are proposed to predict the sway of such frames. Schueller’s equation, originally proposed for rigid frames, has also been modified to incorporate the effect of connection stiffness and proposals are made to add a new term to the equation. All the proposed methods are explained using worked examples, with results being compared against the FE results. Comparisons show that

References (17)

There are more references available in the full text version of this article.

Cited by (11)

  • Seismic performance of semi-rigid steel-concrete composite frames

    2020, Structures
    Citation Excerpt :

    Calculation of effective length factor for the columns [16–18] Evaluation of storey drifts [19–21] Evaluation of minimum connection stiffness and strength required for satisfying the seismic drift limitations [20,21]

  • Optimum design of semi-rigid connections using metamodels

    2012, Journal of Constructional Steel Research
  • Review on the modelling of joint behaviour in steel frames

    2011, Journal of Constructional Steel Research
    Citation Excerpt :

    Other international codes of practice which also consider joint behaviour are those of the USA in AISC-ASD [29], LRFD [30], AISC-ASD/LRFD [31] and China in GB [32]. There is currently a great range of studies of steel frames with semi-rigid connections: Jaspart [33], Jaspart and Maquoi [34], Weynand et al. [35], Chen [36], Braham and Jaspart [37], Ashraf et al. [38], Cabrero and Bayo [39], Bayo et al. [40], da S. Vellasco et al. [41], Ashraf et al. [42], Yang and Lee [43], Faella et al. [44], da Silva et al. [45], Daniũnas and Urbonas [46], Sekulovic and Nefovska-Danilovic [47], Bel Hadj Ali et al. [48], Ihaddoudène et al. [49], Mehrabian et al. [50], Darío [51], etc. These studies were concerned with two principal themes [52]: (1) the evaluation of the mechanical properties of the joints in terms of rotational stiffness, moment resistance and rotation capacity, and (2) the analysis and design procedures for frames including rotational joint behaviour.

  • Sway of semi-rigid steel frames, part 2: Irregular frames

    2007, Engineering Structures
    Citation Excerpt :

    Figs. 11 and 12 present two such cases where flexibility factors for two different stepped frames with one tall bay are compared to their constituent regular frames. Flexibility factors for multi-bay frames for use in Eq. (1) are easily obtainable using the design charts reported in [2] except for 1 bay multi-storey regular frames. Hence FE models were developed for 1 bay semi-rigid regular frames following the same technique reported in [2] to generate design charts for such frames; these are given in Fig. 13.

View all citing articles on Scopus
View full text