AbstractAn investigation into the behaviour of pressed-in piles has been conducted. The press-in method of pile installation allows large pre-formed foundation piles to be constructed without the noise and vibration associated with conventional dynamic techniques, and with minimal requirement for temporary works. This investigation is divided into two parts; a fundamental study of the mechanics of press-in pile installation in sand and a sequence of field tests to examine the behaviour of pressed-in piles at full scale. The mechanics of pile installation have been studied using a plane strain calibration chamber. A new system for deformation measurement in plane strain modelling and other forms of geotechnical testing has been developed. This system combines techniques of digital photography, Particle Image Velocimetry (PIV) and close range photogrammetry. A series of validation experiments demonstrated that the system offers greater accuracy and precision than existing measurement techniques. This improved performance is achieved concurrent with an order-of-magnitude increase in the number of measurement points that can be established within the observed soil. A series of 8 calibration chamber tests is reported. The pattern of soil displacement during pile installation was measured. These measurements were of sufficient quality to allow soil strain paths during installation to be calculated. The influence of soil type and initial density was examined, and the post-installation strain distribution was found. The concentration of shear and volumetric strain close to the pile tip was quantified, and a reversal of strain direction as the soil passes around the pile shoulder was observed. A zone of highly compacted soil was observed immediately below the pile tip and along the pile shaft. Contraction of this sleeve of broken soil grains was observed with continued penetration of the pile. A mechanism is proposed to link this kinematic observation to the distribution of shaft friction close to the tip of displacement piles. A further mechanism is proposed to predict the distribution of external shaft friction along the upper part of a pile shaft. This mechanism is based on vertical arching theory, and is an extension of a previous approach for the prediction of internal shaft friction. Four series of field tests using pressed-in piles were conducted. The first series demonstrated that internal shaft friction is well predicted by vertical arching theory. Since vertical arching evolves according to an exponential function, pile performance can be dramatically influenced by only small changes to the governing parameters. The improvement of driveability using an internal driving shoe was investigated. The final series of load tests demonstrated a novel foundation solution in which the high shaft friction created by vertical arching can be 'switched on' after installation. This is achieved using a construction sequence involving H-section piles. During press-in installation of each pile, the geometry does not create vertical arching. During loading of the entire structure, arching occurs. This leads to a high positive group effect, and an efficient foundation structure. Keywords: pile, sand, foundation, press-in, particle image velocimetry, photogrammetry, geotechnical.