Piezoelectric stack actuator life test

Future NASA interferometer missions require actuators for precision positioning to accuracies of the order of nanometers. For this purpose, commercially available multilayer piezoelectric stack actuators are being considered for driving these precision positioning mechanisms. These mechanisms have potential mission operational requirements that exceed 5 years and the nominal actuator requirements for the most critical actuators on the these missions were estimated from the Modulation Optics Mechanism (MOM) and Pathlength control Optics Mechanism (POM) mechanisms which were developed for the Space Interferometry Mission (SIM). At a nominal drive frequency of two hundred and fifty hertz one mission life is calculated to be 40 Billion cycles. In order to test the feasibility of using these commercial actuators for these applications and to determine the reliability and the redundancy requirements of these actuators a life test study was undertaken. In this study a set of commercial PZT stacks configured in a potential actuator flight configuration (pre-stressed and bonded in flexures) were tested for up to 100 billion cycles. The test flexures allowed for two stacks to be mechanically connected in series. The tests were controlled using an automated Lab-View control and data acquisition system that set up the test parameters and monitored the waveform of the stack electrical current and voltage. The samples were driven between 0 and 20 Volts at 2000Hz to accelerate the life test and mimic the voltage expected to be applied to the stacks during operation. During the life test the primary stack was driven while the redundant stack was open circuited. The stroke determined from a strain gauge and the temperature and humidity in the chamber and the temperature of each individual stack were recorded. In addition other properties of the stacks were measured at specific intervals. These measurements included the displacement from a Capacitance gap sensor and impedance spectra. The degradation in the stroke over the life test was found to be small (<3%) for the primary stacks and estimated to be < 4% for the redundant stacks. It was noted that about half the stroke reduction occurred within the first 10 billion cycles. At the end of the life test it was found that by applying DC voltage levels (100 V) above the life test voltage we could initially recover about half of the lost stroke with again some degradation in the long term. The data up to 100 billion cycles for these tests and the analysis of the experimental results will be presented in this paper.
^1,2