EXPERIMENTAL STUDY ON SEISMIC PERFORMANCE EVALUATION OF TRADITIONAL TIMBER FRAME STRUCTURES WITH LARGE HANGING WALLS

Abstract

 In Japan, there are a lot of traditional timber buildings which fit the history, the culture and the environment in each area. In our investigation of traditional timber houses in the important district of groups of historic buildings, mud walls which have various specifications and crosspieces called Nuki and structures which consist of tall hanging walls and beams called Sashigamoi (called large hanging walls in this study) were found. Timber frame structures with hanging walls are very important factor to think of seismic safety of timber buildings because they can lead to collapse of the whole buildings by breakage of columns at joints of the lower end of hanging walls. Although static loading tests and shaking table tests of timber frames with hanging walls have been conducted, the failure behavior and deformation performance have not been analyzed in detail.
 As the current methods of seismic capacity evaluation, “Implementation Guidance for Basic Seismic Assessment of Important Cultural Properties (Buildings)” proposed by Agency for Cultural Affairs and “Seismic Evaluation and Retrofit Methods of Wooden Houses” proposed by the Japan Building Disaster Prevention Association have been used. Although the consistency between the current methods and the test results has been examined, the hanging walls of the specimens are not tall and the analysis is not enough about shear force and deformation performance of the columns, flexural strength of the columns and limit deformation evaluation of the frame structures with hanging walls.
 Therefore, the purpose of this study is demonstrative elucidation of dynamic characteristics of timber frame structures with large hanging walls in order to construct a reasonable and practical seismic evaluation method. First, we conduct static loading tests of timber frame structures with full walls or large hanging walls. The height of the full wall specimens and the large hanging wall specimens is 2.70m and 3.87m, respectively. The span length of the specimens is 1.82m. The height of hanging walls is 1.80m. The specification of walls and the number of spans are experimental variables. Second, we analyze damage states and mechanical characteristics of the specimens and their elements such as mud wall, columns and Sashigamoi. Finally, we reveal the applicability and problems of the current methods based on the test results.
 The major findings obtained from the research are summarized as follows:
 a) Different specifications of Nuki do not make much difference of the hysteresis characteristics of the full wall specimens in spite of the different failure states of walls.
 b) The hysteresis characteristics of the large hanging walls can be calculated using the hysteresis characteristics of the full wall specimens which have the same specification of wall.
 c) Timber frames with hanging walls do not lose the restoring force immediately after breakage of columns because the broken columns have the restoring force. Within this study, rotational angle of the specimens is more than about 1/15 rad when the restoring force decreases to 80% of the maximum value.
 d) Shear force of a column is not depend on only the sum of half of the distances from the column to the adjacent columns on the both sides, because shear force is affected by the sectional performance and Young's modulus of the column, the location of the column against the loading direction and breakage of the surrounding columns. In addition, we indicated that it is need to consider rotational deformation of columns at column-Sashigamoi joints in evaluation of deformation of timber frames with hanging walls. From the above, it is revealed that there is room for improvement on the current methods in seismic performance evaluation of timber frames with large hanging walls.