Effects of Welding Method, Loading Rate, and Loading Sequence on Fracture Behavior of Welded Steel Beam-to-Column Connections

本研究的目的是找出在1994年的北岭地震和1995年的晓谷根-南部地震中如此多的焊接梁-柱连接断裂的原因。研究的重点是焊接方法、加载速率和加载顺序，并进行了以下步骤的研究：(1)收集以往的试验数据，建立梁柱焊接连接数据库；(2)对历次地震的破坏模式进行调查；(3)以焊接方式、加载速率和加载顺序为参数的梁柱连接试验；(4)对所得结果进行了综合，并提出了新的设计方案。主要调查结果总结如下：(a)建立了一个数据库，其中载有以前对梁与柱的焊接连接进行的试验。发现连接的延性能力非常分散。(b)对损坏的梁柱连接处进行的调查显示，最终导致断裂的裂缝更多是在焊接通道孔的趾部、径流片、焊接尖端或焊接金属中开始的。(c)考察了不同类型的焊缝接入孔，并量化了其对连接延性的影响。研究发现，与常规细部相比，减少孔尺寸的修改细部可以确保更大的延性能力，其延性能力是常规细部的2到3倍。(d)大地震动作用下框架的数值分析表明，梁端承受的应变速率在0.2 ~ 0.3/s之间。(e)与以往的推测相反，动态加载比准静态加载表现出更大的延性能力，且断裂模式更具延性。在动态加载过程中发现显著的温升是这种失效模式的可能原因。(f)根据焊接孔的类型、预期延性能力和失效模式，提供了焊接梁柱连接的抗震设计准则。少

The objective of this research is to identify the reasons why so many welded beam-to-column connections fractured in the 1994 Northridge and 1995 Hyogoken-Nanbu Earthquakes. The focuses of the investigation are the welding method, loading rate, and loading sequence, and a study comprising the following steps was carried out : (1) collection of previous test data and development of database on welded beam-to-column connections ; (2) survey of damage patterns exhibited in the previous earthquakes ; (3) tests of beam-to-column connections with the welding method, loading rate, and loading sequence as parameters ; and (4) synthesis of obtained results and proposals for new design procedures. The major findings are summarized as follows :(a) A database that contains previous tests of welded beam-to-column connections was developed. The ductility capacity of the connections was found to scatter very significantly.(b) A survey on damaged beam-to-column connections revealed that cracks that ev … More entually caused fractures were initiated in one of the toe of the weld access hole, run-off tab, weld tip, or weld metal.(c) Various types of weld access holes were examined, and their effects on the connection ductility capacity were quantified. It was found that modified details in which the size of the holes was reduced can ensure greater ductility capacity by a factor of 2 to 3 compared to the conventional details.(d) Numerical analysis for frames subjected to large ground motions showed that the strain rate sustained by the beam end would be in a range of 0.2 to 0.3/s.(e) Contrary to previous speculation, dynamic loading exhibited larger ductility capacity than quasi-static loading, with the mode of fracture being more ductile. Significant temperature rise during dynamic loading was found to be a likely cause for this mode of failure.(f) Guidelines were provided as to the seismic design of welded beam-to-column connections in terms of the type of weld access holes, expected ductility capacity, and mode of failures. Less