Technical Report Influence of the geometric and material characteristics on the strength of glued joints made in chestnut timber Dolores Otero Chans * , Javier Estévez Cimadevila, Emilio Martín Gutiérrez Department of Construction Technology, Escuela Técnica Superior de Arquitectura, University of A Coruña, Spain article info Article history: Received 18 February 2008 Accepted 18 June 2008 Available online 27 June 2008 abstract This paper summarizes the experimental campaign carried out on joints made with threaded steel rods glued with epoxy in sawn hardwood timber. This study deals with a very little-known field of the glued joints in timber, because previous studies focus on the analysis of joints made with glued laminated tim- ber (glulam) obtained principally from softwood. A total of 180 specimens were tested with the aim of evaluating the incidence of several geometric and material parameters in the axial strength of the joints. The experimental results were compared with the existent design criteria. The research results question the application of the existent design proposals on joints made with hardwood, and they stress the need to delve deeply into the experimental study of glued joints made with sawn timber of high density. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction The joints made with glued-in bars are an option of great inter- est for the design of wooden structures. Some of their advantages are the aesthetic benefits associated with the hidden design of these joints and their capability to carry out rigid joints with a high transmission of efforts from within the wooden pieces. These advantages have also turned the glued joints into a valu- able system for recuperation and rehabilitation of ancient wooden structures. These joints are specially advantageous to elements that have suffered severe attacks of xylophagous organisms. One of the main limiting factors of these joints is the fact that there are no standards which can guarantee and make easy the de- sign attending to some designing simple parameters. Since the 1980s several experimental studies aimed at evaluating the strength of glued joints in timber have been carried out, which derived in various design proposals of empiric origin. The Eurocode 5 has included a design proposal [1], as an informative annex. Anyway, the existent experimental studies are mainly limited to the field of the structures made of glulam, manufactured in soft- wood. There are hardly any experimental studies with glued joints in hardwood, in spite of the fact that this wooden type is the most used in the structures of historic buildings. An extensive experimental campaign aim to evaluate the strength of glued joints made in hardwood have been carried out for our research team, at the University of A Coruña, during the last years [2]. The obtained results have been also used to evaluate the correspondence with the previous existent studies on glulam [3]. The experimental results of an extensive experimental research on glued joints made in chestnut hardwood timber are described in this paper. 2. Experimental program 2.1. Materials In order to realize joints tests that can provide a guideline for historical and new structures, a usual sawn timber was selected. The choice of the timber species was made bearing in mind the enormous potential that the glued joints offer in the field of resto- ration and rehabilitation of historic and old buildings, being this the most promising use in our country. Considering that glued joints are very suitable for rehabilitation, a hardwood species tim- ber used in traditional edification in Galicia, chestnut timber (Castanea sativa Mill.), was tested in the experimental campaign. In Spain there is no graded wood of chestnut because there are only standards of classification in resistance grade for softwood timber, poplar and eucalyptus (this is a habitual problem in the use of hardwood for structural use). Due to the non-existence of strength graded sawn timber, wood obtained from solid timber boards of a thickness of 7 and 10 cm was used. A study of the de- fects of the pieces (small knots or cracks) was made for making possible the evaluation of their potential repercussion in the pos- terior failure mode of the specimens. The more habitual defects of the tested timber were knots and cracks, always with small size, which affected in some way up to 40% of specimens. In any case, specimens which had defects with significant size were discarded. All the tested specimens were previously measured and weighed, and their moisture content got registered with an elec- tronic two-prongs moisture-meter (Protimeter TimberMaster). 0261-3069/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.matdes.2008.06.041 * Corresponding author. Tel.: +34 981 167 000x5282; fax: +34 981 167 051. E-mail addresses: marilo@udc.es, marilooch@gmail.com (D. Otero Chans). Materials and Design 30 (2009) 1325–1332 Contents lists available at ScienceDirect Materials and Design journal homepage: www.elsevier.com/locate/matdes