No abstract

<p>Nanometer-resolution position sensing is a critical requirement in several precision applications such as Atomic Force Microscopy (AFM), especially over large bandwidths. Giant magnetoresistance-based (GMR) sen?sors have shown great promise for this; however, they exhibit significantly higher 1/f noise even over large band?widths, which limits the achievable SNR. Therefore, the sensing system needs to be designed at the system-level in order to minimize excess intrinsic noise generation and transmission, as well as external noise pickup. This article, both analytically and experimentally, investigates the noise characteristics of the GMR sensor and its readout circuit components, and their effect on SNR. A general process for designing a low-noise high-bandwidth readout circuit, applicable not just to GMR sensors but also to other high?1/f-noise sensors, is presented. A resolution of 2.5 nm over a bandwidth of 100 kHz is demonstrated on an AFM nanopositioner, a ten-fold improvement over previously reported performance and comparable to other state-of?the-art, but more complex, position sensing schemes. The readout scheme is simple to implement using COTS com?ponents and easily extendable to higher bandwidths, unlike other schemes, such as modulation/demodulation, where the requirement for increasingly higher carrier frequencies renders further improvements in bandwidth impractical.  </p>

<p>  Nanometer-resolution position sensing is a critical requirement in several precision applications such as Atomic Force Microscopy (AFM), especially over large bandwidths. Giant magnetoresistance-based (GMR) sen?sors have shown great promise for this; however, they exhibit significantly higher 1/f noise even over large band?widths, which limits the achievable SNR. Therefore, the sensing system needs to be designed at the system-level in order to minimize excess intrinsic noise generation and transmission, as well as external noise pickup. This article, both analytically and experimentally, investigates the noise characteristics of the GMR sensor and its readout circuit components, and their effect on SNR. A general process for designing a low-noise high-bandwidth readout circuit, applicable not just to GMR sensors but also to other high?1/f-noise sensors, is presented. A resolution of 2.5 nm over a bandwidth of 100 kHz is demonstrated on an AFM nanopositioner, a ten-fold improvement over previously reported performance and comparable to other state-of?the-art, but more complex, position sensing schemes. The readout scheme is simple to implement using COTS com?ponents and easily extendable to higher bandwidths, unlike other schemes, such as modulation/demodulation, where the requirement for increasingly higher carrier frequencies renders further improvements in bandwidth impractical.  </p>

<p>  Nanometer-resolution position sensing is a critical requirement in several precision applications such as Atomic Force Microscopy (AFM), especially over large bandwidths. Giant magnetoresistance-based (GMR) sen?sors have shown great promise for this; however, they exhibit significantly higher 1/f noise even over large band?widths, which limits the achievable SNR. Therefore, the sensing system needs to be designed at the system-level in order to minimize excess intrinsic noise generation and transmission, as well as external noise pickup. This article, both analytically and experimentally, investigates the noise characteristics of the GMR sensor and its readout circuit components, and their effect on SNR. A general process for designing a low-noise high-bandwidth readout circuit, applicable not just to GMR sensors but also to other high?1/f-noise sensors, is presented. A resolution of 2.5 nm over a bandwidth of 100 kHz is demonstrated on an AFM nanopositioner, a ten-fold improvement over previously reported performance and comparable to other state-of?the-art, but more complex, position sensing schemes. The readout scheme is simple to implement using COTS com?ponents and easily extendable to higher bandwidths, unlike other schemes, such as modulation/demodulation, where the requirement for increasingly higher carrier frequencies renders further improvements in bandwidth impractical.  </p>

An analytical methodology is presented for the systematic design of a serial kinematic flexure-guided high speed nanopositioning scanner. Approximate relations for the first natural frequency in different directions, achievable range, and in-plane cross-coupling between the axes are obtained considering each stage as a simple one-dimensional mass spring system. Parametric studies are performed to compare these characteristics for a particular range of flexure dimensions. The robustness of the design to manufacturing tolerances, and especially their influence on cross-coupling is investigated. The relations obtained are experimentally verified on a single-axis test system and the measured natural frequencies closely match analytical and FEA predictions. A two-axis nanopositioning system is designed, which has the fast and slow scanning axes’ resonances at 26.36 kHz and 5.28 kHz. The design is validated using finite elements, and the predicted actuation-direction resonances (25.5 kHz and 5.24 kHz respectively) closely agree with those found analytically.