Development of system-on-chip cmos-mems pressure sensors
- Autores: Saoni Banerji
- Directores de la Tesis: Jordi Madrenas Boadas (dir. tes.), Daniel Fernández Martínez (dir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2017
- Idioma: español
- Tribunal Calificador de la Tesis: Pascal Nouet (presid.), Joan Pons Nin (secret.), Núria Barniol i Beumala (voc.)
- Programa de doctorado: Programa de Doctorado en Ingeniería Electrónica por la Universidad de las Illes Balears y la Universidad Politécnica de Catalunya
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
The advent of monolithically integrated CMOS-MEMS resonant pressure sensors has paved the way for integration of sensors and electronics in a single chip. The significance of MEMS integrated with CMOS stems from their many advantages, i.e. reduced cost, small size, continuous miniaturization, high performance, and reliability while reasonably maintaining their key performance parameters. The main objective of this thesis is to conduct research on the design and characterization of CMOS MEMS resonant pressure sensors based on the Quality factor (Q) measurement, and to design a system-on-chip that is capable of estimating its performance on a digital platform. This work intends to enable the integration of sensors together with signal conditioning VLSI circuits on a single chip with minimum space, weight and energy requirements targeted to the implementation of autonomous systems. Maintaining the quality of the received signal and implementation of the sensor without intensifying the overall system cost and complexity is projected to be the primary purpose for assimilating electronics with the pressure sensor.
The thesis initially entails the design optimization and validation of a CMOS-MEMS resonant pressure sensor with an enhanced sensitivity at atmospheric pressure. The sensing principle of a resonant pressure sensor is based on the variation of the quality factor with pressure. The designed square plate pressure sensor was manufactured in 250-nm CMOS technology after verifying the optimization procedure by finite element modeling. With the in-house BEOL metal-layer release, this sensor can be monolithically embedded in the same substrate as standard CMOS integrated circuits, resulting in a significant cost and area reduction. Secondly, comprehensive characterization results of the CMOS-MEMS resonant pressure sensor were presented. The experimental measurements of the manufactured resonant pressure sensor were found to have good convergence with that of the analytical and finite element modeling to validate the optimization procedures undertaken to design the device. The sensitivity of the manufactured device is approximately -0.09 at atmospheric pressure and increases to -0.3 at 40 kPa i.e. in the lower pressures of slip flow regime. Over the full-scale pressure range of 0.1 to 100 kPa and a temperature range of ¿10 °C to 85 °C, Q from 450 to 62.6 have been obtained. Besides, static variations of the device capacitance have been measured and analyzed with temperature to evaluate the spring softening and the pull-in effects. A nonlinearity analysis was performed to assess the device stability. Furthermore, a statistical mismatch analysis was carried out to determine the deviation of resonance with etching time and ascertain maximum device yield.
Thirdly, a behavioral modeling technique for CMOS-MEMS micro resonators is presented that enables simulation of a MEMS resonator model in Analog Hardware Description Language (AHDL) format within a system-level circuit simulation. The resonator was modeled into Verilog-A code and successfully simulated within Cadence framework. Analysis has shown that simulation results of the reported model are in agreement with the device characterization results. As an application of the proposed methodology, simulation and results of the model together with an integrated monolithic low-noise amplifier is exemplified for detecting the position change of the resonator. This has finally been followed by the design of a signal processing chain interfaced with the prototype sensor that illustrates the extraction of system quality factor to a digital domain.
