Ultra-Low Noise Mechanical Frequency-Divided MEMS-Based Oscillator

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

This research program aims to harness new capabilities in miniature mechanics that enable portable personal radar with resolution commensurate with much larger radars used on demanding platforms, such as planes, boats, and tracking systems. Here, because the resolution of any radar depends heavily on the stability of its interrogating signal, the quality of the internal oscillator that generates this signal is among the most important determinants of radar performance. It is for this reason that the most capable radars use frequency signal generators based on 'Poseidon' oscillators that derive their stability from bulky (and expensive) sapphire-loaded cavity resonators. However, the size and cost of this approach restricts it to use in large non-portable applications. The research herein breaks this paradigm by recognizing that an oscillator referenced to an ensemble of tiny mechanically resonant structures constructed in diamond material can potentially equal and perhaps outperform a Poseidon oscillator, all in a substantially smaller form factor. Indeed, sizes small enough to enable high-resolution personal radar might be within reach, and this could enable a host of new personal capabilities, including ranging, remote location services, and local area sensing to identify opportunities, e.g., sales on interesting items, favorite foods, etc., as one walks past them. The research specifically explores parametric mechanical frequency division using micromechanical resonators and sub-20-nanometer electrode-to-resonator gaps that together propel the short-term stability and figure-of-merit of oscillators towards that of the best radar oscillators, but in substantially smaller size. Here, tiny electrode-to-resonator gaps enabled by advances in nanofabrication technology increase the electromechanical coupling of capacitive-gap transduced micromechanical resonators to ranges that support gigaHertz oscillators, all while retaining their record-setting quality factors, both of which drive down phase noise and power consumption. Sub-20-nanometer gaps also permit wider frequency tuning, as well as the use of parametric excitation at higher resonance frequencies, which in turn enables mechanical approaches to reducing phase noise further via a combination of frequency division and close-to-carrier filtering, all with nearly no power penalty. While gap reduction enables unprecedented micromechanical resonator and oscillator performance, it also elucidates fundamental physical limitations expected to manifest more prominently at the nano-scale, including nonlinear noise multiplication, undue injection locking, Casimir forces, enhanced (unwanted) acceleration sensitivity, and less resilience against finite fabrication tolerances and stress. While possibly detrimental to performance, these physical limitations open fertile ground for scientific study. Indeed, lessons learned by modeling the influence of nonlinearity on oscillator limiting and phase noise are likely key to propelling micromechanical oscillators into the stated higher end applications. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.