Nonlinear Coda Wave Interferometry in concrete - effects of pump conditioning

This dataset has been collected on a laboratory basis. Two concrete samples of identical dimensions (50 cm × 25 cm × 12 cm) and composition (CEM I 52.5 N cement, 0/4 and 4/12 calcareous sand and aggregates, water-to-cement ratio of 0.55) were used. One of the blocks has been subject to a 3-point bending test to generate a single crack in the middle, which will be referred to as the damaged specimen or block here. The other block is intact and will be called the undamaged block or specimen.

In fact, in typical NCWI experiments, a pump-probe approach is used where a high-amplitude low-frequency wave, called the pump wave, is applied to temporarily and reversibly activate material nonlinearities, which are then probed by a low-amplitude high-frequency probe wave. The pump waves are frequency-swept chirps transmitted continuously at a fixed amplitude for each pump level, and this amplitude is increased step-by-step. On the other hand, the coda probe emitter sends shorter chirps, asynchronously at a certain repetition rate. The probe signal is then received via the coda transducers. The theory of Coda Wave Interferometry (CWI) is used to process probe signals. More specifically, the stretching technique is utilized to estimate relative velocity changes of the coda signals.

In Materials exhibiting elastic nonlinearity, the changes in pump amplitude are expected to cause measurable changes in the measured CWI parameters, which must be more pronounced in a damaged sample. The mesoscopic hysteretic nonlinearity theory explains this, where the behavior is attributed to the brick-and-mortar structure of such materials, where bonds open/close or shear. As a result of this nonlinear behavior, in addition to classical and non-classical nonlinearity terms, the pump excitation causes a phenomenon called non-equilibrium dynamics, which is a stiffness loss during conditioning caused by the pump excitation and a slow log-time recovery once the excitation stops. According to this phenomenon, replacing the systematic standard NCWI pumping procedure with a less-predictable, more chaotic excitation like in-service loading, such as traffic on a bridge, makes interpreting the results more complicated.

Therefore, the dataset in this document is collected to study the effects of ‘pumping duration,’ specifically, as a step toward this overall objective. The results have already been published in a paper in JASA. This dataset is presented as a representative of the experiments performed for the mentioned study. More specifically, this dataset is related to Experiment 1 in this paper: