SQUID-based setup for the absolute measurement of the Earth's magnetic field

T Schönau, M Schmelz, V Zakosarenko, R Stolz, M Meyer, S Anders, L Fritzsch, H-G Meyer
2013 Superconductors Science and Technology  
HP54). This Highlight was submitted by the authors of the highlighted paper which appeared in SUST 26 (2013) 035013. The authors noted that the paper generated significant interest (Editors comment). SQUID-based magnetic sensors exhibit outstanding sensitivity and bandwidth. But, due to their periodic voltage-flux characteristic, they are not suited for absolute magnetometry. Furthermore, their application in unshielded environment is challenging, because magnetic transients may interrupt the
more » ... ux locking loop (FLL) and introduce step-like shifts into the output signal that cannot always be corrected by postprocessing of the acquired data. The sinusoidal-like voltage-flux characteristic of a SQUID is unique (and thus suited for absolute magnetometry), if the absolute value of the magnetic field strength B is below a certain threshold value B max , which depends on the effective area A of the SQUID: (1) To achieve both, a large B max and a good sensitivity, we propose to use a cascaded SQUID setup as depicted in Figure 1 . It consists of several coplanar SQUID magnetometers with largely different effective pickup areas that are integrated on a single chip. Because of their spatial proximity and equal orientation, each SQUID of the cascade measures approximately the same magnetic field strength. The effective area of the smallest SQUID (herein named reference magnetometer) is about A R ≈ 0.05  0 /µT, which corresponds to B max ≈ 10 µT. The absolute output of the reference SQUID is now used to calculate the branch of the voltage-flux characteristic on which the next SQUID in the cascade (the intermediate SQUID) is locked, resulting in a more precise absolute value of the measured magnetic field component since the effective area of two consecutive SQUIDs in the cascade increases by about two orders of magnitude. This principle is repeated up to the most sensitive SQUID, which determines the noise level of the final measurement value. The white noise level of the sensitive SQUID in our setup is about B n,s = 6 fT/Hz 1/2 . The dynamic range of the system, defined as the ratio of the maximum peakto-peak signal amplitude to the achievable signal resolution limited by the noise B n,s in a 1 Hz bandwidth, is about 190 dB. The value of B max ≈ 10 µT is actually too small for absolute magnetometry in the Earth's magnetic field (B Earth ≈ 50 µT). But for many applications, a redesign of the reference SQUID is not necessarily needed. If the reference SQUID is not locked on the correct branch, the error is n 0 /A R ≈ n20 µT, where n is an unknown integer. This quantized offset can quite easily be estimated prior to the measurement, for example by a rotation of the setup or a by measuring the offset inside a simple magnetic shielding.
doi:10.1088/0953-2048/26/3/035013 fatcat:iooeon4np5a4dkkotqmqb5qvsi