Characterization and control of surface parallel mirror actuators

The Surface Parallel Actuator (SPA) mirror testbed consists of an aluminum off-axis paraboloidal (OAP) mirror. The mirror backplate has a triangular/hexagonal rib structure with 42 rib cutouts for insertion of the SPA’s. This mirror development was a pathfinder task to investigate the development of SPA mirrors using Commercial Off The Shelf (COTS) piezoelectric stacks and to uncover potential issues of this approach. In the current testbed designed for room temperature testing we have used piezoelectric stack actuators from PI Ceramic (PICMA® PZT stacks P-885.51) preloaded with monolithic bi-elliptical flexures (Al 7075) manufactured using electric discharge machining (EDM). The stacks were instrumented with strain gauges to provide closed loop control of the SPAs in the mirror ribs and to investigate the accuracy of open loop control algorithms. This paper is a continuation of the characterization and control developments that we have previously published about the performance of flexure piezoelectric multilayer stack actuators embedded in the mirror [Sherrit et. al. 2022a, 2022b, 2024a, 2024b]. Although the SPA testbed was designed for room temperature operation we have been cognizant of a desire to operate down to 100K (near IR) in future iterations and have investigated properties down to 100K when appropriate in terms of cost and difficulty. In the initial work [Sherrit et. al. 2022a] we measured the free stack properties between 300 K to 100 K including the stroke vs voltage curves, the electric displacement vs field curves, the thermal strains, the Coefficient of Thermal Expansion (CTE) under different electrical boundary conditions and the stroke creep. The bi-elliptical flexure was designed to preload the stacks to keep the piezoelectric under compression even when driven in extension. In the subsequent research [2022b] we investigated the stroke of the piezoelectric stack in the flexures and the response of the actuators embedded in the mirror [2024a]. In order to investigate a potential screening test for the actuators we also investigated the dynamic response of the stack/flexures up to 100 kHz [2024b]. In this paper we summarize the testing of the free stacks, flexured stacks and the flexured stacks while populated in the mirror and report on additional thermal measurements including stroke stability under close loop control and thermal stability of the stroke and the change in curvature at 100 K as determined by strain gauges mounted on the actuators and mirror analogs. We also report on our initial testing of the influence functions to enable WaveFront Error (WFE) control. In the initial theoretical SPA studies for this mirror configuration we determined using the NASTRAN DEFORM card to produce the influence functions (IFs) for 1 μm displacements. Using generated Z4 to Z15 Zernike mode errors and applying corrections using the IFs of all the actuators it calculated a theoretical 15 times reduction in the RMS WFE assuming linear control. To measure the actual IFs for each actuator position, we developed control electronics that can power, address, and control all 42 SPAs individually. Using an interferometer, we measured the mirror surface shape created by each individual actuator over a range of voltages. Initial WFE control suggests RMS reductions of the order of 10 times base on the measured influence functions. During the development there were many lessons learned, and the final part of this paper will focus on how these issues can guide the next generation of actuated mirrors through some of these development pitfalls, including the potential to separate the optical and actuator planes and the use of flex circuitry to minimize wiring and harness complexities.