EXPERIMENTAL INVESTIGATION OF YIELD C-DEVICES UNDER MONOTONIC AND CYCLIC LOADING

  Share :          
  62

Laboratory testing is the cornerstone for validating numerical and analytical models in structural engineering, particularly when studying the energy-dissipation capacity of metallic yielding devices. Although many experiments have been carried out on other yielding devices, experimental studies dealing specifically with C-shaped yielding devices (C-devices) are limited. These devices had not previously been tested in isolation under cyclic loading, and the effects of laser cutting and heat treatment on the properties of their steel had not been studied. Chapter 7 of the thesis therefore experimentally investigates the influence of the dissipation-zone length (Ld) and the circular transition radius (r) on the device's energy-dissipation capacity and deformation tolerance, while also verifying the accuracy of the numerical models developed in Chapters 5 and 6. The experiments covered four geometric device types (R30-1hb, R30-2hb, R50-1hb and R50-2hb). They were laser-cut from 10 mm thick S275 mild steel plate, with a circular transition radius of 30 or 50 mm and a dissipation-zone length of 58 or 116 mm, representing roughly one third of the expected full-scale device. A total of 16 devices were tested, eight under monotonic loading and eight under cyclic loading. In addition, a few specimens with poor detailing, that is, with sharp corners or a right-angle (90°) internal corner and cut from the same plate, were made to demonstrate that neglecting the detailing of critical corners leads to brittle behaviour owing to stress concentration. To fix the devices in the testing machine, steel fixtures made of S355 steel were designed, using plates thicker than the device plate to prevent them from yielding when the device reached its maximum load. They were connected to the device with high-strength bolts. An INSTRON machine with a 500 kN capacity was used, together with two data-acquisition systems recording up to 17 channels per test. Displacements were measured with five linear variable differential transformers (LVDTs), and strains with strain gauges attached to both edges of the dissipation zone. In addition, Digital Image Correlation (DIC) with a digital camera was used to determine the average strain in the target region, after a random paint speckle pattern had been sprayed onto it. The loading protocol included monotonic tests to determine the yield force (Py), yield displacement (Δy) and elastic stiffness (K), and reversed cyclic tests in which displacement amplitudes were increased up to 12 times the yield displacement, the limit permitted by the 20 mm capacity of the displacement-control system. These were applied at a slow strain rate of about 2.5×10⁻⁶ per second to avoid strain-rate effects. A second displacement history, with lower amplitudes and more cycles, was used to validate the numerical model implemented in SAP2000. The monotonic tests showed that the effect of the circular transition begins to appear when the material enters the strain-hardening stage, and that the elastic region is governed mainly by the length of the dissipation zone. The yield force ranged between 35 and 43 kN depending on the device type. The increase in yield force for heat-treated devices was slight (about 2.7%), compared with a difference of roughly 12.19% for untreated devices. The residual deformations were concentrated within the dissipation zone after the end of the test. In the cyclic tests, all devices exhibited stable hysteretic behaviour, with no degradation in stiffness or strength up to displacement amplitudes of 12 times the yield displacement, confirming their reliability as energy dissipators. Resistance was higher as the dissipation-zone length became shorter: R30-1hb recorded the highest initial stiffness, the highest maximum force (54 kN) and the greatest maximum curvature, while R30-2hb recorded the lowest values. The maximum tensile force reached between 1.18 and 1.54 times the yield force owing to strain hardening, and the parameters of the Wen model were determined from the results for use in representing the device's cyclic behaviour. The strain-gauge readings showed that the dissipation zone behaved inelastically and that the neutral axis lies about 21 mm from the outer edge of the device. When the experimental results were compared with the numerical models, the adopted SAP2000 model proved able to predict the device's cyclic response with an acceptable level of agreement and with closely matching hysteresis loop shapes, although its loading and unloading stiffness was slightly higher than in the tests. The larger tensile force observed in the tests is attributed to friction, which was not included in this model. By contrast, the ADINA model reproduced the deformed shape and the inelastic stress distribution in the dissipation zone with good accuracy, and the strain maps obtained by DIC matched the model's predictions, confirming the expected yielding mechanism and the concentration of energy dissipation in this zone. The study concludes that the device's yield force is affected by the dissipation-zone length and the circular transition radius, particularly in longer devices, and that increasing these two variables reduces the average maximum curvature. C-devices designed with proper detailing show stable and reliable cyclic behaviour, whereas neglecting corner detailing leads to early brittle failure. The results also confirmed the reliability of the adopted finite element models, which supports the use of these devices as an economical and effective solution for the seismic retrofitting of precast reinforced concrete structures. Prepared by: Dr. Oday Al-Mamoori Specialization: Civil Engineering – Structural Engineering and Earthquake Engineering Source: Al-Mamoori, O. H. H. (2019). Seismic Redesign of Precast Portal RC Frames using Yield C-devices (Chapter 7: Experimental investigation of the C-device under monotonic and cyclic loading). PhD thesis, Civil Engineering Department, School of Environment and Technology, University of Brighton, United Kingdom.