Seismic retrofit of two-story earthen historic buildings using steel plates

Abstract

The Spaniard colonizers of Latin America built constructions using earthen materials, and consequently, adobe and rammed earth became part of the cultural heritage of northern South America. However, these heritage structures have been deteriorating due to the inherent structural vulnerability, mainly caused by earthquakes and moisture. Previous earthquakes have shown that unreinforced earthen buildings are susceptible to seismic damage, resulting in heritage, economic, and human losses. In recent years, research has focused on seismic retrofitting alternatives that included confining wooden/steel elements, steel cables, tie beams, steel tensioners, internal reinforcements, mesh reinforcements, polypropylene bands, etc. Confinement of earthen walls with steel strips was proposed recently and consisted of structural steel plates (101.6 mm width and 6.35 mm thickness) installed on the inner and outer faces of earthen walls (vertically and horizontally) forming a grid. The A36 steel plates are separated a distance between 1000 mm and 1500 mm and are joined with through bolts spaced every 500 mm. Despite the advances in the steel plate retrofitting technique for earthen buildings, design equations are not available, and studies on two-story full-scale specimens are limited, particularly in Colombia. This is an important issue because a large percentage of historic earthen buildings in Colombia have two stories. Therefore, the effectiveness of retrofitting techniques for these cases is not well understood. To address this gap in knowledge, this research project tested large-scale two-story walls or segments retrofitted with steel plates using pseudo-static and dynamic (shake table test) load protocols in one and two directions, respectively. Despite the high variability involved in the mechanical properties of earthen materials and the uncertainty in the interaction between the steel plates and the rammed earth, an analytical model, and an empirical design equation (validated with experimental tests) were proposed to predict the flexural out-of-plane strength of reinforced earthen elements. Additionally, a full-scale two-story rammed-earth wall with openings (0.65 m thickness, 6.20 m height, and 5.95 m length) was subjected to cyclic in-plane shear loads with two MTS actuators. These actuators applied loads until the reparable damage threshold was reached, which corresponds to the point at which the resistance shown in the hysteresis loops began to decrease. Later, the wall was reinforced with steel plates, and the same test protocol was repeated. The experimental results suggested that the in-plane stiffness was restored, and the lateral load capacity was increased by about 200 % on average. Furthermore, the unreinforced wall had a drift capacity of 0.5 %, while the reinforced wall reached a maximum drift of 1.8 %. Finally, two 1:2 scale rammed earth walls (unreinforced and reinforced with steel strips) were tested on a bi-axial shaking table device (X and Y ground motions). Based on the experimental tests, the unreinforced wall presented irreparable damage and high residual drifts at acceleration levels corresponding to a return period of 475 years (PGAy=0.43g). In contrast, the reinforced wall had a better seismic performance with lower damage levels and was highly resilient, withstanding an earthquake greater than the one with a 2500 return period without collapsing (PGAy=0.76g). Therefore, the experimental research conducted suggests that the steel plates reinforcement technique improves the seismic performance of two-story heritage rammed-earth walls while reducing damage and protecting human lives. Confinement with steel plates also reduced early failures and permanent residual drifts, allowed a higher range of non-linear displacements, reduced cracks and damage, and improved out-of-plane stability.