FREQUENCE RESPONSE The operating frequency of an actuator may depend on beam’s geometry/size, thermal boundary conditions, etc. It has been reported that micro thermal actuators could operate at high frequencies, possibly exceeding 1 KHz [2]. To characterize the frequency response of the actuator accurately, the displacement of the shuttle and the excitation of the actuator needed to be sampled at a frequency at least twenty times faster than the excitation frequency. A Phantom high speed camera system with A/D input channels was integrated with a microscope. The displacement of the shuttle in high resolution videos and the excitation signal in discrete numbers were recorded in sync. The quantitative displacement data were determined from the videos in post-experiment image analyses. The integrated system is shown in Figure 8. The excitations were Haversine functions V =Vo(1-cos(2 πf))/2, where Vo was the amplitude of the input voltage and f was the frequency. Figure 9 shows the normalized steady-state excitation V/Vo and displacement d/do of the actuator at two different frequencies, f = 0.1 and 5.0 Hz. The value of do was the static displacement corresponding to Vo, which was determined from the characteristic displacement-voltage curve of the actuator shown in Figure 4(b). The image and data acquisition rate was 24 fps for 0.1 Hz input and 200 fps for 5.0 Hz input. For the examples shown in Figure 9, Vo = 4.0 volt and do = 95 um. At the frequency f = 0.1 Hz, shown in Figure 9(a), (d/do)max = 1 and (d/do)min = 0. The shuttle has a full range of stroke, the same as that of static loading, but with a small amount of phase lag. At the frequency f = 5.0 Hz, (d/do)max = 0.54 and (d/do)min = 0.20, shown in Figure 9(b). The stroke is significantly reduced, only 34% of the stroke of static loading; it also has a large phase delay of about 75 degree. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 d/do V/Vo 0.1Hz 2.0Hz 5.0Hz 10Hz 20Hz 50Hz Figure 10 Normalized displacement-voltage curves for Vo = 4.0 V at various frequencies. A number of normalized displacement-voltage curves of different frequencies, but same Vo = 4.0 V, are plotted in Figure 10. As the frequency gets higher, the range of the stroke becomes smaller and the stroke approaches the asymptote d/do = 0.38 or d = 36 um. The high frequency response of the actuator to excitation Vo = 4.0 V is equivalent to the static loading at 2.4 V. LOAD-DISPLACEMENT CURVE To measure the force of the actuator, an external load cell was utilized. Shown in Figure 11(a), the load cell is mounted on a micro manipulator with a probe attached on the other side. The end of probe is bent about 90o, so it can reach the shuttle and keep the load cell and shuttle aligned on the same axis, Figure 11(b). Under the microscope, the probe can be positioned and aligned at the end of the shuttle by controlling x, y, z and up-down of the manipulator, Figure 11(c). Then the probe is ready to load the actuator using the x-control. Since the bent probe is compliant, the displacement of the manipulator is not the same as the shuttle. The displacement of the shuttle is determined from the digital microscope. The load-displacement curve of the actuator was characterized under the constant excitation voltage. During experiment, the actuator was first activated by a constant excitation Vo and the shuttle moved freely to a displacement do. Then in each following loading steps, the probe was moved slowly to push the shuttle for a distance δ. The shuttle’s total displacement became d = do - δ. The values of the force F and displacement d were recorded. It was repeated until snap-thru occurred or being stopped manually. 207
RkJQdWJsaXNoZXIy MTMzNzEzMQ==