Modeling-free evaluation of resonant liquid sensors for measuring viscosity and density

A performance estimation method for resonant viscosity and density sensors is described. The ultimate sensor precision and the separability of density and viscosity from measured frequency spectra are estimated. The approach relies on using recently introduced models which are flexible enough to describe a large class of sensors. It is therefore not required to develop a specific sensor model and a parameter extraction method for each newly devised sensor. With the method introduced in this contribution, it is sufficient to adjust and validate a general model using at least three test liquids. Then, the error propagation and the error bounds for viscosity and density can be estimated and consequently the required signal-to-noise ratio (SNR) or the amount of required data points can be determined. This method enables direct comparison of different sensor setups as will be demonstrated for piezoelectric tuning fork sensors, electrodynamic platelet sensors and steel tuning forks.