Sensors (Dec 2021)
The Computational Fluid Dynamics (CFD) Analysis of the Pressure Sensor Used in Pulse-Operated Low-Pressure Gas-Phase Solenoid Valve Measurements
Abstract
This paper presents a flow analysis of the original pressure sensor used to determine times until full opening and closing of the pulse-operated low-pressure gas-phase solenoid valve. The sensor in question, due to the fast variation of the process lasting several milliseconds, has high requirements in terms of response time and ability to identify characteristic parameters. A CFD code has been employed to successfully model the flow behavior of the original pressure sensor used to determine times until full opening and closing of the pulse-operated low-pressure gas-phase solenoid valve at different inlet flow conditions, using the Eulerian multiphase model, established on the Euler–Euler approach, implemented in the commercial CFD package ANSYS Fluent. The results of the modelling were validated against the experimental data and also give more comprehensive information on the flow, such as the plunger displacement waveform. The flow calculations were dynamic in nature; therefore, the experimental plunger displacement waveforms were entered as input in the software for dynamic mash implementation. In identifying the times until full opening and closing, the characteristic points of the pressure waveform on the pressure sensor plate were adopted. CFD flow calculations confirmed the accuracy of identifying the times until full opening and closing by relating them to the results from the plunger displacement sensor. The validation of the results of calculations with the analyzed sensor and the original stand also confirmed the correctness of the use of this type of method for the assessment of gas injector operating times. In the case of time until full opening, the CFD calculations were shown to be consistent with experimental tests, with only a few cases where the relative difference with respect to the displacement sensor reached 3%. The situation was slightly worse in the case of time until full closing, where the results of CFD calculations were in agreement with the displacement sensor, while the experimental test stands had a relative difference of up to 21%. It should be remembered that the sensor evaluates times below 5 × 10−3 s, and its construction and response time determine the use depending on the adopted level of accuracy.
Keywords