Feedback control of MEMS devices has the potential to significantly improve device performance and reliability. One of the main obstacles to its broader use is the small number of on-chip sensing options available to MEMS designers. A method of using integrated piezoresistive sensing is proposed and demonstrated as another option. Integrated piezoresistive sensing utilizes the inherent piezoresistive property of polycrystalline silicon from which many MEMS devices are fabricated. As compliant MEMS structure's flex to perform their functions, their resistance changes. That resistance change can be used to transduce the structures' deflection into an electrical signal. The piezoresistive microdisplacement transducer (PMT) is a demonstration structure that uses integrated piezoresistive sensing to monitor the output displacement of a thermomechanical inplane microactuator (TIM). Using the PMT as a feedback sensor for closed-loop control of the TIM provided excellent tracking with no evident steady-state error, maintained the positioning resolution to ±29 nm or less, and increased the robustness of the system such that it was insensitive to significant damage.
Feedback control has proven useful in improving reliability and performance for a variety of systems. However there has been limited success implementing feedback control on surface micromachined MEMS devices. The inherent difficulties in sensing microscale phenomena complicate the development of an economical transducer that can accurately monitor the states of a surface micromachined system. We have demonstrated a simple and effective sensing strategy that uses the piezoresistive property of the polysilicon thin film of which surface micromachined MEMS devices are fabricated. The states of the device are monitored by measuring the change in resistance of flexible members which deflect as the device moves. Measurement of the output displacement of an in-plane thermal actuator is presented as a candidate application. While there still is a noise issue to be dealt with, this approach provides adequate signal strength to implement feedback control using off-chip analog circuitry. Implementation of proportional/integral control on the system is successfully demonstrated.
Compliant piezoresistive MEMS sensors exhibit great promise for improved on-chip sensing. As compliant sensors may experience complex loads, their design and implementation require a greater understanding of the piezoresistive effect of polysilicon in bending and combined loads. This paper presents experimental results showing the piezoresistive effect for these complex loads. Several n-type polysilicon test structures, fabricated in MUMPs and SUMMiT processes, were tested. Results show that, while tensile stresses cause a linear decrease in resistance, bending stresses induce a nonlinear rise in resistance, contrary to the effect predicted by linear models. In addition, tensile, compressive, and bending loads combine in their effects on resistance. The experimental data illustrate the inability of linear piezoresistance models to predict the piezoresistive trends of polysilicon in bending and combined loads, indicating the need for more complete nonlinear models appropriate for these loading conditions.
Utilizing the piezoresistive properties of polysilicon as an on-chip sensing mechanism facilitates the implementation of feedback control on surface-micromachined MEMS devices. We have performed nanopositioning resolution tests on a MEMS thermal actuator, both open and closed loop, to demonstrate the performance improvements possible with feedback control. A thermomechanical in-plane microactuator (TIM), fabricated using the MUMPS fabrication process, was used in this study. The actuator was coupled to a piezoresistive displacement sensor (PRDS) that was fabricated as part of the same process. Measurements of the actuator output, taken using a scanning electron microscope, show that nanopositioning repeatability improved from ±59 nm to ±31 nm when feedback control is employed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.