Research Academics  
 
   
 

2D-Arrays of MEMS Surface Hot-Wires for Wall Shear-Stess
Measurement in Turbulent Flows

 
Contact Person: Dr.-Ing. Ha-Duong Ngo

 

Introduction

In the last few years, active flow control has become a focus of the work of researchers throughout the world. Active flow control can help to reduce the level of flow-related noise, to increase the lift of airfoils in flight phases like take-off and landing where the flow is susceptible to separation, and to decrease the fuel consumption of airplanes by reducing the overall drag of the airplane.To achieve effective active flow control a system of sensors for detecting the current state of the flow is needed. The more accurate the sensors outputs are, the more likely an appropriate reaction can be deduced. Furthermore, a high temporal an spatial resolution of the sensor system is necessary to obtain precise information about fast changing and small-sized flow phenomena typical for turbulent flows.The most promising measurement concept to gain high-frequency information to characterise a turbulent flow is the indirect measurement of wall shear-stress by using a surface hot-wire. MEMS processes can help to reduce the size of surface hot-wires and thus allow the realisation of a sensor system with high spatial resolution by combining many single MEMS surface hot-wires to build a 2D-array of such sensors.

Structure and Sensor Design

Starting point for the realisation of 2D-arrays of MEMS hot-wires is a thoroughly optimised single MEMS surface hot-wire. The structure of such a sensor is depicted in figure 1: a wire of 2 µm height and width is connected to conducting paths, which keep the wire strained. The wire itself is situated over a cavity in the substrate material and thus thermally well isolated. Polyimide foil is used as substrate material, allowing both single MEMS sensors and 2D-arrays to be fitted to curved surfaces like airfoils or turbine blades.

Figure 1: Structure of a single MEMS surface hot-wire on a polyimide foil

The dimensions of the cavity and the hot-wire have been optimised by thermal FEM analyses. Single MEMS sensors and MEMS line-arrays have been realised using sputtering, lithography, wet chemical and reactive ion etching and are shown in figures 2 and 3.

>   

Figure 2: Single MEMS surface hot-wire.        Figure 3: Line-array of five MEMS surface hot-wires.

Measurement Results

The realised MEMS surface hot-wires show an electrical resistance that is well comparable to values calculated from the specific electrical resistance of the hot-wire material and the geometry of the wires. The average temperature coefficient of electrical resistance, TCR, is 4.5*10-3 1/K, a high value compared to the TCR of other MEMS hot-wires. Calibration results of the hot-wire signal against the readings of a conventional wall shear-stress balance at an overheat ratio of 1.5 are depicted in figure 4 and show the typical characteristics of a hot-wire calibration curve. The cut-off frequency of the MEMS hot-wires has been determined by using a sine-signal-sweep and is higher than 15 kHz.

Figure 4: Calibration of a MEMS surface hot-wire

Conclusion

It has been shown that a MEMS surface hot-wire suspended over a cavity at its ends can be realised without any mechanically supporting structures like beams, membranes or bridges. Furthermore, polyimide foil has been used as substrate material, thus providing a high mechanical flexibility of single sensors and 2D-arrays.

Based on the successfully developed single sensors and line arrays, two dimensional arrays of MEMS surface hot-wires will be realised.