Design and Experiment of Anti-Particle-Contamination MEMS Thermal Flow Sensor by Channel Geometry Optimization and...

Micro-electro-mechanical system (MEMS) thermal flow sensors are attractive for compact gas-flow measurement because of their small size, rapid response and low power consumption. However, under non-clean operating conditions, particles carried by the gas can enter the sensing region and progressively deposit on the chip, altering the local flow field and heat-transfer boundary conditions and thereby causing output drift and cumulative measurement error. To improve long-term stability in particle-laden environments, a MEMS thermal flow sensor was designed and implemented, and contamination was controlled along the particle pathway from transport toward the sensing region to deposition on the chip surface. The finite element method (FEM) simulation showed effective thermal isolation between the heating and temperature-sensing regions at ΔT = 20 °C, while the heating power increased monotonically with flow velocity in agreement with heat-loss theory. Particle-laden experiments indicated that the average measurement error was reduced by approximately 22.5% using the curved bypass channel for the flow speed up to 70 SLPM, with a more pronounced improvement at medium and high flow rates. For particles that still reached the vicinity of the chip, the surface electric field was further adjusted through the operating voltage. The sensor was tested under particle-filled flow for 2 h, and the results demonstrate that the contaminated regions were 14.34%, 10.40%, and 3.84% at supply voltages of 7, 8, and 9 V, respectively, indicating that lower operating voltage mitigated particle deposition.