Proceedings Article | 2 August 2014
KEYWORDS: Sensors, Field programmable gate arrays, Light emitting diodes, Stars, Computing systems, Computer simulations, Solar system, Photometry, Clouds, Signal processing
The Whipple mission was a proposal submitted to the NASA Discovery AO in 2010 to study the solid bodies of the Kuiper Belt and Oort Cloud via a blind occultation survey. Though not accepted for flight, the proposal was awarded funding for technology development.
Detecting a significant number of Trans Neptunian Objects (TNOs) via a blind occultation survey requires a low noise, wide field of view, multi object differential photometer. The light curve decrement is typically a few percent over timescales of tenths of seconds or seconds for Kuiper Belt and Oort cloud objects, respectively. To obtain a statistically interesting number of detections, this photometer needs to observe many thousands of stars over several years since the rate of occultation for a single star given the space density of the TNOs is low. The light curves from these stars must be monitored with a sensor with a temporal resolution of rv 25-50 ms and with a read noise of< 20 e- rms. Since these requirements are outside the capability of CCDs, the Whipple mission intends to use Teledyne H2RG HyViSI Silicon Hybrid CMOS detectors operating in "window" read mode. The full Whipple focal plane consists of a 3x3 array of these sensors, with each sensor comprised of 1024x 1024 36/μm pixels. Combined with the telescope optic, the Whipple focal plane provides
a FOV of rv36 deg2 . In operation, each HyViSI detector, coupled to a Teledyne SIDECAR ASIC,
monitors the flux from 650 stars at 40 Hz. The ASIC digitizes the data at the required cadence and an FPGA provides preliminary occultation event selection.
The proposed 2010 Whipple mission utilized a spacecraft in a a "drift-away" orbit which signifi cantly limited the available telemetry data rate. Most of the light curve processing is required to be on-board the satellite so only candidate occultation events are telemetered to the ground. Occul tation light curves must be processed in real time on the satellite by an Field Programmable Gate Array (FPGA). A simple, real time band pass filter, called the Equivalent Width (EW) algorithm, has been instantiated in the FPGA. This EW filter selects for telemetry only those occultation event light curves that differed significantly from noise.
As part of our technology development program, a key facet of the proposed Whipple focal plane was constructed and operated in our laboratory consisting of a single HyViSI H2RG sensor, a Teledyne SIDECAR ASIC, and a flight-like Virtex-5 FPGA. In order to fully demonstrate the capabilities of this photometer, we also made a occultation light-curve simulator. The entire system can generate simulated occultation light curves, project them onto an H2RG sensor, read out the sensor in windowing mode at 40 Hz, pass the data to an FPGA that continuously monitors the
light curves and dumps candidate occultation events to our simulated Ground Support Equipment (GSE). In this paper, we summarize the technical capabilities of our system, present sample data, and discuss how this system will be used to support our proposal effort for the next Discovery round.
