Dependence of signal-to-noise ratio on excitation current and signal frequency in spin valve sensors

Spin valves are widely used as magnetic microsensors. Their performance is limited by the presence of noise. In this article, signal and noise responses of microfabricated spin valve sensors are investigated in the linear region and in the nonlinear region close to saturation where peaks in noise output have been reported. The anisotropy of the free layer and the pinned layer are in a cross configuration. The noise of the sensor is of 1/f type at low frequency. The signal-to-noise ratio (SNR) as a function of signal frequency was measured near saturation. It is found that below 300 Hz the SNR increases almost linearly as the signal frequency increases. The variation of SNR with excitation current, measured at different frequencies, is shown. The SNR shows a nonlinear characteristic. At low frequency the increase in signal power is matched by that in noise. The SNR curve at 35 Hz shows a maximum as the current increases. The slope of other curves increases with increasing frequency. We explain these curves in a model with thermal excitation. (C) 2003 American Institute of Physics.