Fracture Tests and Finite Element Analysis of Diaphragm Connection in Steel Beam-to-Box Column Joints

Steel box columns are widely used in steel building structures in Taiwan because of the strong axes in two directions. In order to transfer the beam end moment to the column, diaphragm plates of the same thickness and elevations as the beam flanges are usually welded inside the box column. The electro-slag welding (ESW) process is usually used in attaching the diaphragms to the column flanges. This ESW process has been widely used in steel beam-to-box column joints in Taiwan because of its’ convenience and efficiency. However, the ESW may increase the hardness of the welds and heat affected zones (HAZs), while reduce the Charpy-V Notch (CVN) strength in the HAZ. This situation could cause the diaphragm to column flange weld to suffer premature fracture before a large plastic rotation is developed in beam-to-box column joints. Results of 22 full scale steel beam-to-box column connection tests conducted in 2008 by Lin et al. show that the probability of ESW fracture at the HAZs adjacent to the ESW is about 18%. Cheng et al. continued the study in 2011 with new tests on full scale beam-to-box column subassembly and ESW components. Test results show that the fractures of the diaphragm-to-column flange welds are sensitive to the eccentricity between the elevations of the ESW and the beam flange. In order to quantify the critical eccentricity and the effectiveness of predicting the fractures, this study adopts the fracture prediction model proposed by Kanvinde in 2004 and utilizes finite element model (FEM) analysis to correlate the test results. In this study, two additional beam-to-box column connection subassembly tests have been conducted with a different loading protocol or the shape of ESW chamber. It is found that the fractures of the aforementioned HAZs can be reasonably well predicted using the proposed procedures on applying the fracture model. Test results show that under the standard and increasing loading protocols, the instances of the HAZ fractures in the specimens correspond to the total beam end deformations of the 2nd and 1st cycles of 0.03 radian, respectively. It is confirmed that the fracture instances can be predicted based on the cumulative plastic deformation in the HAZs. Under the two same standard loading protocol, the flared ESW chamber alter the fracture location from the HAZ in the diaphragm weld to the beam flange weld, and delay the fracture at the 2nd cycle of 0.03 radian to the 1st cycle of 0.04 radian. Tests confirm that the possible fracture of the diaphragm to column flange welds can be mitigated by enlarging the chamber of the ESW. When the fracture prediction model is applied, the material parameters were firstly established from the Circumferential Notched Tensile (CNT) tests and FEM analysis. Subsequently, these parameters were used to predict the fractures observed in the ESW component tests and beam-to-box column connection subassembly tests. The fracture locations and instances can be reasonably well predicted by a suitable FEM model analysis. Thus, the effectiveness of CNT and the fracture model are confirmed. Analytical results also show fracture instances and locations are sensitive to the relative locations of the ESW and the beam flange. Thus, the importance of ultrasonic test in assuring the quality of the ESW is evident. Finally, a parametric study on the eccentricity between the ESW and beam flange locations is conducted using 45 models with various vertical and horizontal eccentricities of ESW with respect to the beam flange. The parametric study results show that the size of critical zone of ESW is indeed the most important factor affecting the fracture instance. When the relative vertical eccentricity between ESW and beam flange is increased, the fracture instances is accelerated. Enlarging the critical zone of ESW effectively reduces the possibility of fracture as evidenced by using a flared shape ESW chamber in the aforementioned test specimen.