Among SOMTC’s world class facilities is the testbed for the Phased Array Mirror Extendible Large Aperture (PAMELA). The PAMELA is a 0.5-meter aperture Cassegrain telescope comprised of 36 hexagonal primary mirror segments. Each mirror segment is 7 cm flat-to-flat, is spherical, and has a radius of curvature of 1.52 meters. The full-aperture telescope has a f/# of 1.5. A catadioptric secondary mirror and some tertiary collimating optics compensate the inherent spherical aberration.

The PAMELA originated at Kaman Aerospace as a SDIO/DARPA program in the 1980’s.When Kaman’s funding dried up before the project was completed, MSFC acquired the PAMELA hardware in 1993 and completed the integration and testing task under the MSFC Center Director’s Discretionary Fund (CDDF).

Each PAMELA mirror segment has three degrees of freedom: tip, tilt and piston. Voice-coil actuators are mounted on the backs of the segments as control effectors. A 36-subaperture Shack-Hartmann wavefront sensor provides the local tip and tilt information for each segment. Inductive coil edge sensors provide relative piston information. Wavefront sensor and edge sensor outputs are sampled at 5 kHz and processed by a stack of digital signal processors (DSP).

Control system processing is done locally on the DSPs,and at 5 kHz the DSPs output the position commands to the actuators. In August 1994 complete tip/tilt/piston closed-loop control was successfully demonstrated on the entire 36-segment array. [view PAMELA Optical Table Layout]

Although stable tip/tilt/piston control was successfully demonstrated, the steady-state jitter remaining in the system prohibited PAMELA performance from approaching the diffraction limit. Several sources contributed to the residual jitter. They included the lively voice-coil actuators, which are pulsed at 5 kHz, and the noisy lateral effect diodes (LED) in the Shack-Hartmann wavefront sensor. Unstable air currents and temperature fluctuations also affected the steady-state wavefront error. Several steps were taken to remedy the situation. First, the wavefront sensor was replaced with detectors and electronics that had a noise floor more than an order of magnitude lower that the old LEDs. Second, the actuators were retrofitted with tiny viscoelastic damping devices. Combined, these two actions reduced steady-state vibration amplitudes by two orders of magnitude. Baffling the optical path helped reduce the effects of air currents.

In September 1998 phasing a cluster of five adjacent segments was demonstrated on the PAMELA testbed using a Helium-Neon laser source. An intensity hill-climbing approach devised by The Sirius Group of Huntsville, AL was used to phase the segment cluster. In the hill-climbing approach, the five segments were first aligned individually in tip and tilt and held to the reference position via the Shack-Hartmann sensor. Next, one at a time, segments were stepped in piston until the total image intensity hit a local maximum. Shown below is the progression in intensity maximization.

• Phasing the PAMELA Telescope at MSFC
• Image Sensor Spot Intensity Profile

In its current state, the PAMELA is truly a world-class adaptive optics testbed. The PAMELA testbed at MSFC is one of only a handful of places in the world where control of tip/tilt/piston and phasing of segmented primary mirrors has been successfully demonstrated. Current research at the PAMELA includes investigations into image-plane based wavefront sensing, including phase retrieval, hill-climbing, and hierarchical processes trading off sequential versus parallel segment phasing. MSFC is also exploiting panoramic imaging systems for sensing edge misalignments of segmented mirrors and figure shapes of continuous monolith mirrors. The PAMELA testbed is also a center for technology transfer spinoffs. Lessons learned from the PAMELA have been transferred to the NGST project and to McDonald Observatory’s Hobby-Eberly Telescope.