Ironless Inductive Position Sensor for Harsh Magnetic Environments

Linear Variable Differential Transformers (LVDTs) are widely used for high-precision and high-accuracy linear position sensing in harsh environments, such as the LHC collimators at CERN. These sensors guarantee theoretically infinite resolution and long lifetimes thanks to contactless sensing. Furthermore, they offer very good robustness and ruggedness, as well as micrometer uncertainty over a range of centimeters when proper conditioning techniques are used (such as the three-parameter Sine-Fit algorithm). They can also be suitable for radioactive environments. Nevertheless, an external DC/slowly-varying magnetic field can seriously affect the LVDT reading, leading to position drifts of hundreds of micrometers, often unacceptable in high-accuracy applications. The effect is due to the presence of non-linear ferromagnetic materials in the sensor’s structure. A detailed Finite Element model of an LVDT is first proposed in order to study and characterize the phenomenon. The model itself becomes a powerful design tool for possible countermeasures to the interference effect. In particular, a combination of magnetic shielding and DC polarization is proposed to reduce the drift due to the external field. Nevertheless, such solutions cannot lead to complete immunity, given the unavoidable presence of magnetic materials in the sensor. Taking the CERN application as a starting point, this thesis aims at conceiving, modelling and characterizing a valid alternative to LVDTs for harsh magnetic environments, which would guarantee magnetic-field-immune position sensing while keeping all the advantageous properties of LVDTs. The Ironless Inductive Position Sensor (I2PS) is an air-cored structure made of 5 coaxial coils. The position sensing is achieved by spatially-variable magnetic fluxes, which give rise to position-dependent coil voltages, just as for LVDTs. The complete electromagnetic model of the sensor is proposed, showing the working principle and demonstrating the magnetic-field immunity from a theoretical viewpoint. In addition, a high-frequency electromagnetic analysis is performed, in order to model the skin and proximity effects in the conductors and foresee their impact on the sensor’s functioning. The models are validated with FEM simulations and experimental measurements. The thermal behaviour of the sensor is also investigated and an effective compensation algorithm is proposed to cancel the temperature-dependence of the position reading. In addition, a smart real-time reading algorithm is proposed in order to significantly reduce the estimation error of standard three-parameter Sine-Fit algorithms when an additional sinusoidal signal is present on the main waveform. Finally, a generic optimization procedure is proposed in order to maximize the performances of the sensor in terms of sensitivity. Taking this procedure as a guideline, an actual I2PS optimized prototype is designed and manufactured, having the specifications of the LHC collimators application as a reference. The optimized prototype shows immunity to external ramped and sinusoidal fields, as expected. In addition, it is used for the experimental validation of the models and the reading techniques, which demonstrate their effectiveness.