Numerical investigation and design of fully chord supported tubular T-joints. (15th July 2021)

• Record Type: Journal Article
• Authors: Pandey, Madhup; Chung, Kwok-Fai; Young, Ben
• Is Part Of: Engineering structures. Volume 239(2021)
• Publication Date: 2021-07-15
• Subjects: Cold-formed steel -- Fully chord supported joints -- Design proposals -- FE analysis -- High strength steel -- Tubular joints; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2021.112063
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Numerical investigation was conducted for CFHSS fully chord supported T-joints. A total of 285 finite element analyses was carried out in the parametric study. Chord face, chord sidewall and combination of these two failure modes were observed. Static joint strengths were compared with the predictions of EC3 and CIDECT. New design rules were proposed for CFHSS fully chord supported T-joints. Abstract: A numerical investigation focusing on the design of fully chord supported cold-formed high strength steel (CFHSS) tubular T-joints is presented in this paper. The nominal yield strengths of the tubular members were 900 and 960 MPa. Square, rectangular and circular hollow sections (SHS, RHS and CHS) were used as the brace members, while SHS and RHS were used as the chord members. The tests conducted by Pandey and Young (2019) [1] were used to calibrate finite element (FE) models developed in this numerical investigation. The developed FE models were capable of replicating the test results, including joint resistances, load-deformation histories and failure modes. In order to determine the compression resistance, axial compression loads were applied through the brace members by fully supporting the chord members along their lengths. These calibrated FE models were then subsequently used for the parametric study, which comprised of a total of 189 FE analyses on T-joints with SHS and RHS braces and 96 FE analyses on T-joints with CHS braces. The nominal resistances calculated from the EC3 (2005) [2] and CIDECT (2009) [3] were used to compare the compression resistances obtained from both the tests [1] and the parametric study. In order to check the applicability of the design rules of tubular T-joints, the nominal resistances from these specifications (EC3 (2005) [2] and CIDECT (2009) [3]) were obtained with and without incorporating the recommended reduction (or material) factors. Through the comparison of 307 joints, including 22 tests (Pandey and Young (2019) [1]) and 285 FE analyses, it has been demonstrated that the existing design rules of T-joints given in the EC3 (2005) [2], CIDECT (2009) [3] cannot be directly used for fully chord supported tubular T-joints of S900 and S960 steel grades with validity ranges of governing parameters exceeding the limits specified in these specifications (EC3 (2005) [2], CIDECT (2009) [3]). Hence, new semi-empirical design rules are proposed in this paper for the observed failure modes, including chord face failure, chord side wall failure and their combined failure modes. Reliability analyses were also performed to check the reliability levels of both the existing and proposed design rules.