Search

A database of 26 previous full-scale experiments on rectangular hollow section (RHS) trusses is supplemented by nine tests on a 10-metre-span, simply supported, RHS Warren truss, reported herein. Measured axial forces, bending moments and truss deflections are compared to four 2D, elastic, frame-analysis models consisting of: (i) all joints pinned and concentric; (ii) all joints rigid and concentric; (iii) pin-ended webs connected eccentrically to continuous chords; and (iv) rigid-ended webs connected eccentrically to continuous chords. On average, all four models predict sufficiently accurate axial force distributions and deflections under elastic loading. However, all four models under-predict bending moment magnitudes. Implications for RHS truss analysis according to the 'detailed method' of CSA S16-14 are discussed, and recommendations for modelling are made. Key words: steel structures, hollow structural sections, trusses, static loading, modelling, deflections, connections, frame analysis. Dans cet article, on presente une base de donnees de 26 experiences anterieures, en grandeur reelle, sur des fermes de profile creux rectangulaire (<>), celle-ci etant suppleee a neuf essais sur une ferme de profile creux rectangulaire Warren d'une portee de 10 m, soutenue simplement. Les mesures des poussees axiales, des moments de flexion et des flexions d'une ferme sont comparees a quatre modeles bidimensionnels (2D), elastiques, d'analyse de charpente, qui sont: (i) tous les assemblages goupilles et concentriques; (ii) tous les assemblages rigides et concentriques; (iii) les ames a bouts goupilles reliees excentriquement aux membrures continues; et (iv) les ames a bouts rigides reliees excentriquement aux membrures continues. En moyenne, les quatre modeles prevoient assez precisement les distributions des poussees axiales et les flexions sous chargement elastique. Cependant, les quatre modeles sous-estiment les ampleurs de moment flechissant. Les implications au niveau de l'analyse de fermes de profile creux rectangulaire selon <> de CSA S16-14 sont examinees et des recommandations sont donnees en ce qui concerne la modelisation. [Traduit par la Redaction] Mots-cles: structures d'acier, profiles de charpente creux, fermes, chargement statique, modelisation, flexions, assemblages, analyse de charpente., 1. Introduction Rectangular hollow section (RHS) trusses can be formed by welding together single- or double-mitre cut RHS web members and RHS chords. As shown in Fig. 1, web members [...]

This paper introduces a simple approach to increase flexural strength and ductility of cantilevered thin-walled tubular steel poles, using a partial concrete fill near the fixed base, as an economical alternative to thicker wall tubes that may be necessary to meet the same loading requirements. Because of the moment gradient along the length, only a short length of concrete fill is required and can be optimized. A robust analytical model has been developed to establish the flexural response of a partially concrete-filled steel tube. It accounts for ovalization and local buckling of thin-walled hollow sections as well as plasticity of steel and concrete non-linearity. The model is used to establish the optimal concrete fill length ratio ([L.sub.f(opt)]/L), and was validated using an experimental study carried out on cantilevered steel tubes with various concrete fill lengths. The model was then used in a parametric study to investigate the effects of diameter-to-thickness (D/t) ratio, yield strength ([f.sub.y]) of the tube and concrete strength [f.sub.c')]. It was shown that the [L.sub.f(opt)]/L ratio increases for larger D/t ratios, but is not significantly affected by f or [f.sub.c']. Generally, [L.sub.f(opt)]/L ratio ranged from 0.3 to 0.5 for prismatic tubes with a lateral end load. For a uniformly distributed load, [L.sub.f(opt)]/L ratios are significantly less. The presence of an additional small axial compression load could increase [L.sub.f(opt)]/L ratio, relative to a case of bending only. Key words: pole, tube, thin-walled, concrete-filled, bending, local buckling, ovalization. Cet article presente une approche simple pour accroitre la resistance a la flexion et la ductilite de poteaux d'acier tubulaire a paroi mince en porte-a-faux en utilisant un remplissage partiel par du beton pres de la base fixe en tant que solution economique pour rencontrer les memes exigences de resistance aux charges que les tubes a paroi plus epaisse. Il ne faut du beton de remplissage que sur une courte longueuren raison du gradient de moment sur la longueur, et cette longueur peut etre optimisee. Un modele analytique robuste a ete developpe pour determiner la reponse en flexion d'un tube d'acier partiellement rempli de beton. Ce modele tient compte de l'ovalisation et du flambement local des sections creuses a paroi mince ainsi que de la plasticite de l'acier et la non-linearite du beton. Le modele est utilise pour determiner la longueur optimale de remplissage de beton ([L.sub.f(opt)]/L) et a ete valide au moyen d'une etude experimentale realisee sur des tubes d'acier en porte-a-faux remplis de beton sur differentes longueurs. Il a ensuite ete utilise dans une etude parametrique pour examiner les effets du rapport diametre-epaisseur (D/t), la limite d'elasticite (fy) du tube et la resistance du beton ([f.sub.c']) . Il a ete demontre que le rapport ([L.sub.f(opt)]/L) augmente pour des rapports (D/t) plus eleves, mais il n'est pas affecte de maniere significative par ([F.sub.y])ou ([f.sub.c']). Regle generale, le rapport ([L.sub.f(opt)]/L) variait de 0,3 a 0,5 pour les tubes prismatiques comportant une charge laterale d'extremite. Pour une charge distribuee uniformement, les rapports ([L.sub.f(opt)]/L) sont de beaucoup inferieurs. La presence d'une petite charge en compression axiale additionnelle pourrait augmenter le rapport ([L.sub.f(opt)]/L) par rapport a un cas uniquement en flexion. Mots-cles : poteau, tube, paroi mince, rempli de beton, flexion, flambement local, ovalisation. [Traduit par la Redaction], Introduction Tubular steel poles are commonly used in utility, lighting, telecommunication, and highway signage applications, where they are predominantly subjected to flexure under wind loads. Poles can also be fabricated [...]

For most design and recertification work on tubular structures, it is not practical to determine the joint hot spot stresses experimentally or using detailed finite element analysis. Therefore, parametric equations were developed in the past for various joint geometries to relate stress concentration factors around the joints to basic joint geometrical parameters. Such parametric equations have limited applications to reinforced joints because the nature of reinforcement varies and may be difficult to represent by a generic set of geometric parameters. In this paper, a new method is introduced to include local joint stiffness in the parametric equations represented by a generic characteristic of the joint determined by modal analysis. The parametric equations produced in this paper can be applied to any reinforced T-joint regardless of the nature of reinforcement. This is especially useful in those cases where the exact nature of reinforcement is not known, for example, hidden in the interior or deteriorated through age or fabrication error. The dimensionless joint stiffness parameter can be calculated by a simple modal test and a beam model without needing to know the nature and details of the reinforcement. [DOI: 10.1115/1.2426998] Keywords: generic parametric equation, welded structure, tubular joint, hot spot stress, doubler plate reinforced, stress concentration, fatigue life

This paper presents the results of an experimental investigation of the nondestructive inspection (NDI) of cracks in an intermediate scale tubular T-joint similar to those used in offshore platforms. To simulate the real situation, cracks were generated under fatigue loading. Modal testing and analysis were then carried out to study the vibration characteristics of the tubular T-joints. Strain gages were used as the main transducers to acquire data because strains are more sensitive to the presence of cracks. Strain frequency response functions were particularly analyzed. Three salient phenomena were observed: (a) antiresonance shifts, (b) quasi-static phenomenon, and (c) nonlinearity, all of which are indicative of the presence of cracks, even at a normalized crack size (the ratio of cracking area to load-bearing area) as small as 0.07. Therefore, the method developed in this study renders a more promising NDI technique than the frequency monitoring technique. Finite element analyses were also conducted to confirm the experimental results and to depict the mode shapes.

An experimental research program on nailed steel tubular connections is described. The 'nailing' of one tube to another coaxial tube within it is achieved by a powder-actuated tool and very high strength driven pins (or nails). Combined steel thicknesses of up to 13 mm (1/2 in.) have been easily connected. Furthermore, these structural connections are made very quickly and safely after minimal training with the equipment. A range of tubes up to 406 mm (16 in.) in diameter, with various diameter-to-thickness ratios, tube wall thickness, and lack of fit, have been joined and 42 tube connections have been tested to failure under axial tension or compression loading. The observed failure modes were nail shear failure, tube bearing failure, and tube shear out failure, for which simple prediction formulas are verified. A recommended load and resistance factor design (or limit states design) procedure for such axially loaded tube-in-tube connections is then advocated.

A parametric study of the static strength of circular hollow-section joints with the multiplanar KK configuration, commonly found in both onshore and offshore structures, is presented. The study, carded out using a well-calibrated and validated finite-element model, investigated the effects of the various geometric parameters on static strength. Particular attention was given to joints with [Gamma] ratios, where the amount of experimental data is scarce, to the effect of [Gamma] on strength, and to the influence of the two major failure modes. In all, a total of 50 joints were analyzed. The study has established a clear demarcation boundary between the two failure types as a function of the transverse gap between the braces. A strength database - compiled from the finite-element data generated and 37 existing test results - was used to formulate strength equations through multiple regression analyses. The equations were found to provide strength predictions to a very high degree of accuracy.

A simple analytical expression, amenable to hand calculation, is developed for estimating the ultimate capacity of partially grout-filled damaged tubular members. The analysis is based on the simplifying assumptions that: (1) The bond between the grout and the tube wall could totally fail prior to the attainment of the ultimate capacity; and (2) grout provides sufficient support in the damaged region of the tube and makes the steel cross section fully effective. The load-carrying capacity is then obtained using a beam-column analysis similar in form to that used in the current design codes. The analysis uses the weighted average rigidity of the cross sections (taking into account the effect of the grouted segment) while using the dented section as the critical section. The first yield collapse criterion has been used. Experimental work has been conducted on 10 small-scale specimens. It is shown that the proposed expression provides acceptable lower-bound estimates of the experimental results. A parametric study is performed to establish the relationship between various parameters and their effects on the ultimate strength of partially grout-filled damaged tubes., A model is developed for determining the ultimate strength of partially grout-filled damaged tubular members. The model is based on current design codes for beam-column analysis using the first yield collapse criterion. Analytical and experimental results were compared and the model was found to provide accurate lower-bound estimates of the experimental values. A parametric study was also performed to study the relationships between different parameters and their effects on ultimate strength.

A new modeling technique for the ultimate strength analysis of welded tubular connections subjected to general loading is presented. Based upon the finite element method, the model includes shell and solid elements, large deformations, and elastoplasticity. Unlike other approaches, the ability to simulate the initiation and propagation of fracture is included through the application of continuum damage mechanics (CDM). With this method, cracking is predicted on the basis of a damage variable, and its effect is applied as a reduction in the element stiffness. The development of the material constants is described and numerical results are obtained for two tubular joints that exhibit fracture failure. A comparison between the computed failure loads and those observed experimentally shows a close correlation. With this approach, welded connections that are subjected to arbitrary loading can be efficiently evaluated for strength., A model is developed for the ultimate strength analysis of welded tubular connections under general loading conditions. The model is founded on the finite element method and considers shell and solid elements, large deformations and elastoplasticity. It may be used in estimating crack propagation using continuous damage mechanics. The model is also applied for two welded tubular joints under tensile loading. Good agreement was found between theoretical and experimental results.

A study on geometric and material nonlinear analysis of thin-walled beam columns is discussed. However, the title of the study proves misleading since open sections and the treatment of warping of the cross section are not covered. It is also suggested that the Gaussian quadrature of numerical integration is a better option for evaluating the stifness matrix. Moreover, reduced integration makes the stiffness matrix more flexible, thus hastening convergence. Recommendations for further study are included.