J. Acoustic Emission, 26 (2008) 247 © 2008 Acoustic Emission Group WAVELET ENTROPY AND POWER OF AE SIGNALS AS TOOLS TO EVALUATE DAMAGE IN COATINGS SUBMITTED TO SCRATCH TEST ROSA PIOTRKOWSKI 1,2 , ENRIQUE CASTRO 1 and ANTOLINO GALLEGO 1 1 Departamento de Física Aplicada, Universidad de Granada, 18071 Granada, Spain; 2 ECyT, Universidad Nacional de San Martín, 1650 Pcia. Buenos Aires, Argentina Keywords: Wavelet power; Wavelet entropy; Damage evaluation; Coatings; Galvanized steel; Corrosion. Abstract Wavelet entropy and power were calculated for acoustic emission (AE) signals obtained from scratch tests on hot dip galvanized samples with different corrosion levels. Wavelet power was distributed in different frequency bands, according to damage mechanisms. The frequency bands were automatically obtained by an innovative method that consisted in searching for the relative minima of the wavelet entropy of signals and the subsequent application of a clustering algo- rithm. The damage evaluation entailed studying the evolution of the wavelet power in a specific frequency band, which corresponded to the fracture of the zeta-phase columns of the galvanized coating. Results showed damage to increase along with the level of corrosion. 1. Introduction The early detection of corrosion in galvanized steel used as reinforcement in concrete struc- tures is crucial in monitoring the condition of concrete structures in different environments and under different load circumstances. The identification of damage processes by acoustic emission (AE) has proven very effective in many applications [1-4]. AE through signal processing can be applied on-line as a non-destructive technique on remote or inaccessible parts of a structure. This approach is rooted in the broad field of pattern recognition and damage detection in materials and structures. The present paper is within a program destined to evaluate the adherence of commercial gal- vanized coatings with varying depth, working under different load and corrosion conditions (see [5, 6] and included references). Wavelet transform (WT) was applied to AE signals coming from scratch test (ST) on corroded galvanized steel. The aim was to identify damage mechanisms through the assignment of a few concise and precise parameters to different coating damage processes, and then use some of them to evaluate damage. With this purpose in mind, hot-dip galvanized samples were corroded in a salt chamber and then submitted to ST. Damage mechanisms involving deformation, fracture and/or extraction of the different phases that conformed to the corroded coating, briefly denoted in [5] as oxide, chlo- ride, eta phase and zeta phase, were identified. WT results were adequately obtained as time- frequency AE patterns, which revealed that AE power was distributed in five frequency bands. Consequently, the evolution of AE wavelet power in different frequency bands was carefully compared with SEM and EDX observations along scratches, which permitted the assignment of one mechanism to each band. In particular, signals due to corrosion and non-corrosion mecha- nisms could be easily separated; different mechanisms coexisted along the ST.