[Back]

J. Toledo, V Ruiz-Díez, G. Pfusterschmied, U. Schmid:
"Characterization of oscillator circuits for monitoring the density-viscosity of liquids by means of piezoelectric MEMS microresonators";
Talk: SPIE Microtechnologies 2017, Barcelnoa, Spanien; 05-08-2017 - 05-10-2017; in: "Proc. SPIE 10246, Smart Sensors, Actuators, and MEMS VIII", SPIE, (2017), ISSN: 1996-756x; Paper ID 102461I, 11 pages.

Real-time monitoring of the physical properties of liquids, such as lubricants, is a very important issue for the automotive
industry. For example, contamination of lubricating oil by diesel soot has a significant impact on engine wear. Resonant
microstructures are regarded as a precise and compact solution for tracking the viscosity and density of lubricant oils. In
this work, we report a piezoelectric resonator, designed to resonate with the 4rd order out-of-plane modal vibration,
15-mode, and the interface circuit and calibration process for the monitoring of oil dilution with diesel fuel.
In order to determine the resonance parameters of interest, i.e. resonant frequency and quality factor, an interface circuit
was implemented and included within a closed-loop scheme. Two types of oscillator circuits were tested, a Phase-Locked
Loop based on instrumentation, and a more compact version based on discrete electronics, showing similar resolution.
Another objective of this work is the assessment of a calibration method for piezoelectric MEMS resonators in
simultaneous density and viscosity sensing. An advanced calibration model, based on a Taylor series of the hydrodynamic
function, was established as a suitable method for determining the density and viscosity with the lowest calibration error.
Our results demonstrate the performance of the resonator in different oil samples with viscosities up to 90 mPa·s. At the
highest value, the quality factor measured at 25ºC was around 22. The best resolution obtained was 2.4·10-6 g/ml for the
density and 2.7·10-3 mPa·s for the viscosity, in pure lubricant oil SAE 0W30 at 90ºC. Furthermore, the estimated density
and viscosity values with the MEMS resonator were compared to those obtained with a commercial density-viscosity
meter, reaching a mean calibration error in the best scenario of around 0.08% for the density and 3.8% for the viscosity.

Real-time monitoring of the physical properties of liquids, such as lubricants, is a very important issue for the automotive
industry. For example, contamination of lubricating oil by diesel soot has a significant impact on engine wear. Resonant
microstructures are regarded as a precise and compact solution for tracking the viscosity and density of lubricant oils. In
this work, we report a piezoelectric resonator, designed to resonate with the 4rd order out-of-plane modal vibration,
15-mode, and the interface circuit and calibration process for the monitoring of oil dilution with diesel fuel.
In order to determine the resonance parameters of interest, i.e. resonant frequency and quality factor, an interface circuit
was implemented and included within a closed-loop scheme. Two types of oscillator circuits were tested, a Phase-Locked
Loop based on instrumentation, and a more compact version based on discrete electronics, showing similar resolution.
Another objective of this work is the assessment of a calibration method for piezoelectric MEMS resonators in
simultaneous density and viscosity sensing. An advanced calibration model, based on a Taylor series of the hydrodynamic
function, was established as a suitable method for determining the density and viscosity with the lowest calibration error.
Our results demonstrate the performance of the resonator in different oil samples with viscosities up to 90 mPa·s. At the
highest value, the quality factor measured at 25ºC was around 22. The best resolution obtained was 2.4·10-6 g/ml for the
density and 2.7·10-3 mPa·s for the viscosity, in pure lubricant oil SAE 0W30 at 90ºC. Furthermore, the estimated density
and viscosity values with the MEMS resonator were compared to those obtained with a commercial density-viscosity
meter, reaching a mean calibration error in the best scenario of around 0.08% for the density and 3.8% for the viscosity.

http://dx.doi.org/10.1117/12.2266497