Strength of welded joints under combined shear and out-of-plane bending

Abstract

An experimental and analytical research program was conducted with the objective of investigating the response of welded joints loaded under combined out-of-plane bending and shear. A database of test results, including 60 tests from the University of California in Davis, eight tests from an early research program at the University of Alberta in Edmonton, and 24 tests from Université Laval in Ste.-Foy, was used to evaluate several strength prediction models and the current North American design approaches. This work was complemented by a reliability analysis to assess the level of safety provided by these design approaches. It was determined that both the Canadian Institute of Steel Construction (CISC) and the American Institute of Steel Construction (AISC) approaches provide remarkably conservative predictions of the test results, especially for cases where the welded plate thickness is large. Although a modified version of an approach proposed by earlier investigators in 1972 and based on the method of instantaneous centre of rotation provides an accurate prediction of test results, a simpler strength calculation model that does not require an iterative approach is proposed as a substitute for the current design approaches. The proposed approach provides the desired level of safety for the design of welded joints loaded in shear and out-of-plane bending.

A Boundary Stress Model for Fillet-Welded Connection Plates

“…To account for bearing, either a triangular or rectangular stress distribution in the bearing region is typically considered; see Kwan et al (2010). With reference to Figure 10, which shows a triangular bearing stress distribution, and applying a similar method as before, the resulting location of the neutral axis and nominal flexural strength of the welded boundary is given in Equations 21 and 22, respectively.…”

“…To verify the proposed Design Model 4 with plate bearing, a comparison is made with the test data presented by Kwan et al (2010). The test data presented in the work by Kwan et al reference three primary testing programs in establishing recommendations for a design model: Dawe and Kulak (1972), Beaulieu and Picard (1985), as well as the work conducted by Kwan et al at UC Davis/University of Alberta.…”

“…Of note, Equation 22has been applied to the work of Dawe and Kulak (1972) because they recommended a triangular bearing stress distribution. However, a rectangular bearing stress distribution, Equation 25, is used for the comparison of the work by Kwan et al (2010) and Beaulieu and Picard (1985), as recommended by Kwan et al It should also be noted that Kwan et al primarily presented two types of data-corresponding to specimens fabricated with weldmetal toughness requirements and those without-to reflect the observed influence of toughness on fillet weld ductility. To be as conservative as possible, the test data are presented only for those with no toughness rating; increased weld ductility would allow the design strength of the weld to propagate further along the length of the weld, thus resulting in a larger available strength.…”

“…The average test-to-predicted ratio for the aforementioned test data is 1.20 with a coefficient of variation of 0.19. These values are compared with an average test-to-predicted ratio and coefficient of variation of 1.06 and 0.21, respectively, reported by Kwan et al (2010) between 4.0-4.5 are reported. Thus, because an analysis of the test data with Design Model 4 would result in a larger safety index and because the average test-to-predicted ratio is larger and the variance is smaller, the model is validated as having an acceptable safety index as well.…”

Conjuntos Soldados Submetidos a Carregamento Excêntrico Quase Estático e de Impacto Estrutural “Fora do Plano”

“…Juntas soldadas por filetes são amplamente utilizadas, pois podem alcançar resistências mecânicas relativamente altas e as superfícies necessárias para realizar as soldas são de fácil preparação [1]. Porém, como é usual na soldagem por fusão, principalmente a ductilidade desse tipo de conexão pode ser negativamente alterada devido, entre outros motivos, pelas discrepâncias entre a tensão de escoamento real e a de projeto, além de alterações na resistência ao impacto e natureza do carregamento [2].…”