Regression Modeling of Breakout Strength of an Expansion Anchor Bolt as Influenced by Concrete Aggregates

Abstract

The frictional resistance of an expansion anchor bolt depends directly on the normal forces generated by the anchor expansion mechanism around the anchorage zone of the concrete base material. However, only the concrete aggregates around the anchorage zone have the direct shear contact with the anchor. This study aims to develop a regression model that predicts the breakout strength of an expansion anchor in plain concrete as influenced by the coarse aggregate size, coarse aggregate content, and fine aggregate content. Crushed coarse aggregates of sizes 10mm, 12.5mm, 19mm, and 25mm were used. The coarse aggregate content for each coarse aggregate size was 0.3452, 0.4046, 0.4462, and 0.4750, respectively. The fine aggregate content for each coarse aggregate size was 0.4243, 0.3771, 0.3514, and 0.3341, respectively. A heavy duty type of an expansion anchor bolt was used. There were five samples of base material specimens considered for each size of coarse aggregate. Each specimen was tested first for its compressive strength to ensure that it will meet the designed compressive strength and it will vary only at an acceptable deviation before the installation of the expansion anchors. The installed anchor bolts were then tested for pull-out. Tests results showed that among the parameters considered, coarse aggregate size is the most significant factor that could influence the breakout strength of an expansion anchor and it is optimum at 19mm size. A polynomial regression model is recommended to predict the concrete breakout strength of an expansion anchor bolt as influenced by the coarse aggregate size.

Pull-Out Capacity of Expansion Anchor Bolt as Influenced by Corrosion

“…Previous study supported the results and showed that the pull-out capacity of the expansion stud anchor bolt was affected by other varied factors like concrete aggregate sizes aside from the effective depth of the expansion anchor and the fc' [12]. This study showed that it is also affected by corrosion and that the Neural Network is a powerful tool to predict its pull-out capacity considering the number of days subjected to corrosion, half-cell potential readings, gravimetric test, concrete strength, and waterproofing.…”