Original MEMS wall shear stress sensors developed for separation detection and active flow control on a flap model
Cécile Ghouila-Houri  2, 1, *@  , Ralph Sindjui  2@  , Quentin Gallas  1@  , Eric Garnier  1@  , Alain Merlen  1@  , Romain Viard  3@  , Abdelkrim Talbi  2@  , Philippe Pernod  2@  
2 : Institut d'électronique, de microélectronique et de nanotechnologie  (IEMN)  -  Site web
CNRS : UMR8520, Institut supérieur de l'électronique et du nunérique (ISEN), Université Lille I - Sciences et technologies, Université de Valenciennes et du Hainaut-Cambrésis, Ecole Centrale de Lille
avenue Poincaré, Cité scientifique, BP 69, 59652 Villeneuve d'Ascq cedex -  France
1 : Onera - The French Aerospace Lab  (Lille)  -  Site web
ONERA
F-59045 Lille -  France
3 : Thurmelec
Thurmelec
Fluiditech, Thurmelec, 68840 Pulversheim -  France
* : Auteur correspondant

Active flow control systems are developed to promote air safety, reduce energetic consumption or increase airplanes efficiency. Sensors are needed to measure flow parameters at high frequencies with a high spatial resolution and micro-electro-mechanical system (MEMS) sensors are potential candidates for precise and located measurements.

We present an original MEMS thermal flow sensor designed for flow control and separation detection. The general aim of the work is to run active flow control experiments integrating several MEMS sensors into a motorized deflectable flap model where the actuation is provided by pulsed jets, following previous work performed by Chabert et al. [1]. 

 The micro-sensor is sensitive to the wall shear stress and flow direction [2]. It can be flush mounted to the wall for separation detection and flow control applications. The micro-fabrication process is CMOS-compatible meaning that it allows on-chip integration for designing very small devices.

The micro-sensor structure combines suspended wires, free from the substrate, and micro-bridges used as mechanical supports. Designed to be set perpendicularly to the flow, the sensor presents three parallel heated wires. The central wire is structured with multiple layers with a heater, made of gold, and a sensing wire, placed under the heater and made of Ni/Pt multilayer. This central sensing wire is designed to measure the wall shear stress. The other two sensing wires, placed on both side of the central wire, allow flow direction sensing when considering the temperature variation between them as the wire upstream is more cooled than the wire downstream.

The micro-wires dimensions, 3 µm width for 1 mm length, and the fact that they are suspended over a 20-µm-deep cavity, allow a high gradient of temperature for low power consumption (8 °C/mW for the central wire and 5 °C/mW for the lateral wires).

The micro-sensor was characterized in a turbulent boundary layer wind tunnel by measuring the resistance variations, simultaneously with the wall shear stress fluctuations, measured by near wall hot-wire anemometry. The sensor demonstrates a resistance variation up to 0.3 % for 2.4 Pa for the central wire.

The flow direction measurements were performed using the resistance difference between the two lateral wires. The purpose is to provide a way to detect the presence of a flow separation since in such a situation, the velocity field near the wall is reversed. The sensor setup in the wind tunnel enables to rotate it from 0° to 180°, considering that, at 90°, the wires are parallel to the flow. The results demonstrate the sensor ability to detect the flow direction: the resistance difference is +0.25 Ω for 0°, 0 for 90° and -0.25 Ω for 180°, all for a wall shear stress of 1 Pa.

A second set of experiments were performed by adding an obstacle on the wind tunnel wall, upstream of the MEMS sensor, to cause flow separation at the sensor location. The central wire of the sensor, designed for wall shear stress measurements, present an expected behavior: the sensor detects the decrease of wall shear stress in a separated flow. And the lateral wires detect the flow direction inversion, providing an additional information compared to conventional hot-film sensors.

 The results demonstrate the sensor ability to measure the wall shear stress and to detect flow separation. Currently, the sensors integration in the flap model is in progress and open-loop active flow control experiments will begin at ONERA Lille. The first results will be presented in the full paper.

References:

[1] T. Chabert, Thèse, Université Pierre et Marie Curie - Paris VI, 2014.

[2] C. Ghouila-Houri et al., Appl. Phys. Lett., vol. 109, no. 24, p. 241905, Dec. 2016.


Personnes connectées : 1