The effect of EMAT coil geometry on the Rayleigh wave frequency behaviour

•Frequency content of Rayleigh waves generated by EMATs is measured and calculated.•Observed frequency behaviour is explained using analytical & finite element models.•Method for optimising EMATs for a chosen frequency or wavelength is presented.•Design process moved from ‘trial and error’ to a simple calculation. Understanding of optimal signal generation and frequency content for electromagnetic acoustic transducers (EMATs) is key to improving their design and signal to noise ratio. Linear and meander coil designs are fairly well understood, but other designs such as racetrack or focused coils have recently been proposed. Multiple transmission racetrack coil EMATs, with focused and unfocused designs, were constructed. The optimum driving frequency for maximum detected signal was found to range between 1.1 and 1.4 MHz on aluminium for a 1.5 mm width coil. A simple analytical model based on the instantaneous velocity of a wave predicts a maximum signal at 1.44 MHz. Modelling the detection coil as a spatial square wave agrees with this, and predicts a general relation of fP=0.761v/L between the optimum frequency fP, the wave velocity v, and the coil width L. A time domain model of the detection coil predicts a 1.4–1.5 MHz peak for continuous wave excitation, with a frequency that decreases as the length of the wavepacket is decreased, consistent with the experimental data. Linear coil modelling using the same technique is shown to be consistent with previous work, with improving detection at lower wave frequencies, and signal minima at every integer multiple of the wavelength. Finite Element Analysis (FEA) is used to model the effects of the spatial width of the racetrack generation coil and focused geometry, and no significant difference is found between the focused and the unfocused EMAT response. This highlights the importance of designing the EMAT coil for the correct lift-off and desired frequency of operation.