PSSCT (Pico Satellite Solar Cell Testbed)
The PSSCT mission, a nanosatellite of the STP (Space Test Program) of DoD (USA), is designed to monitor accelerated radiation degradation of triple-junction solar cells in a high radiation environment such as the GTO (GEO Transfer Orbit). Small satellites provide a rapid turn-around platform, or responsive space flight capability, to flight test new solar cells before they are used on major spacecraft.
To minimize overall mission risk, the PSSCT was first placed into a LEO (Low Earth Orbit) flight to test on-orbit the key systems, subsystems, and components such as the I/V curve measuring system, the sun sensors, Earth sensors, rotation rate sensors, and magnetic torque coils in a low-radiation environment. There were no radiation sensors on the LEO flight and instead VGA cameras took their place. The satellite had no closed loop attitude control; hence, it was spun-up about its long axis so that each array would face the sun. 1) 2)
Figure 1: Photo of the LEO PSSCT nanosatellite (left) and its launch tube SSPL (right), image credit: The Aerospace Corporation
The PSSCT nanosatellite was designed and developed at The Aerospace Corporation of El Segundo, CA. The nanosatellite has a box-shaped structure of size 12.5 cm x 12.5 cm x 25 cm with a mass of 6.5 kg. The bus structure has a thick aluminum hull to provide a fairly rigid spacecraft shell and radiation shielding for future high radiation orbits. A wall thickness of 7.5 mm was used to provide a 10 krad interior total dose over a 300 day mission in a 200 km x 35,786 km x 7º GTO orbit. This thick aluminum hull accounts for 50% of the final spacecraft mass. In addition, the highly conductive shell, coupled with high geomagnetic fields at low altitudes, provided a strong eddy-current based rotation damping mechanism for this LEO PSSCT flight.
Each longitudinal side of the spacecraft (Figure 2) has a 4-cell solar array and an attitude sensor composed of a two-axis sun sensor and a single pixel infrared Earth sensor. The sun sensor provides the instantaneous solar inclination angle for the associated solar array.
The +Z side contains electrical contacts to spin up a single momentum wheel to 5000 rpm while it is still inside the launch tube, and four plungers (switches) that turn on the satellite once it has left the launch tube. As the momentum wheel slows down, the spacecraft body begins to rotate about the Z-axis. The Z-axis direction in inertial space was chosen to be perpendicular to the sun-Earth vector and normal to the ecliptic. This allowed the 4 solar arrays to sequentially point towards the sun (at least initially) as the spacecraft rotated.
Figure 2: Illustration of the LEO PSSCT nanosatellite with reference axis and external components (image credit: The Aerospace Corporation)
Inertial sensors: The primary sensors for spin monitoring are ADIS16255 MEMS rate gyros from Analog Devices. These 11.2 mm square by 6.6 mm thick components have a nominal rate range of ± 320º/second, a temperature coefficient of only 25 ppm/ºC, a sensitivity better than 0.1º/s, and an angular random walk of 3.6º/hr1/2 at 25º C. They provide useful rate data up to ± 600º/s. These chips also provide an integrated angle output, but this feature was not used on-orbit. 3)
Three rate gyros are mounted in orthogonal directions on and within the electronics stack located in the center of the LEO PSSCT (Figure 3).
Figure 3: The electronics stack of PSSCT (image credit: The Aerospace Corporation)
The nanosatellite also includes three 2-axis Analog Devices ADIS16003 MEMS accelerometers that are being used to monitor the proper operation of the rate gyros. These accelerometers have a range of ± 1.7 g, a sensitivity of 1.2 mg, and a bias offset of ± 0.14 mg/ºC. 4)
Attitude sensors: PSSCT has three, orthogonal HMC1051ZL single-axis magnetometers from Honeywell. They have a range of ± 6 gauss with a resolution of 120 µGauss; perfectly useful for determining spacecraft orientation using the Earth’s magnetic field in LEO. The sun sensors consist of a 200 µm diameter pinhole mounted above a Hamamatsu S7848 two-dimensional position sensitive detector. This design has a ± 50º angular range in both directions, and an accuracy of ± 2º. A single sun sensor is mounted on the +X, -X, +Y, and –Y spacecraft sides. 5)
The Earth sensors use a Perkin Elmer TPS333 thermopile as an infrared detector over the 5.5 to 20 µm wavelength range. A FOV of 30º is provided on the first pair while a FOV of 50º is given on the second pair of Earth sensors.
Actuation is provided by two momentum wheels with spin vectors aligned along the Z axis. The first wheel was spun up to 5,000 rpm while the satellite was in the SSPL 5510 launcher. The spin-up motor was controlled solely by the Shuttle and had no electrical connections to the PSSCT electronics. A second wheel with a maximum spin rate of 6000 rpm, controllable by the PSSCT, was integrated into the spacecraft to provide angular momentum control about the Z-axis once outside the launch tube.
Magnetic torque coils were also integrated into the PSSCT to test their attitude control authority. They were only used in open loop control. In the future, they will be integrated into a closed loop attitude control system for the GTO PSSCT.
RF communications: The LEO PSSCT uses a modified commercial 915 MHz data radio to provide 38.4 kbit/s communications with the ground station antenna in El Segundo, CA. The spacecraft antenna consists of a single patch antenna designed to provide nearly omnidirectional coverage. This rigid antenna, instead of a deployable flexible antenna, further enhances the structural rigidity of the LEO PSSCT.
Launch and deployment:
PSSCT was a payload of the STS-126 mission of the Space Shuttle Endeavour to ISS (Nov. 14 - 30, 2008). PSSCT was deployed from the Shuttle cargo bay on Nov. 29, 2008 using the SSPL (Space Shuttle Picosatellite Launcher) device.
SSPL was designed by The Aerospace Corporation, El Segundo, CA, and manufactured by Oceaneering Space Systems in Houston, Texas. It is capable of ejecting payloads of size 12.5 cm x 12.5 cm x 25 cm with a mass of up to 7 kg. A spring inside the launcher provides a 1 to 2 m/s relative ejection speed. The SSPL 5510 was first used on STS-116 (Dec. 10-22, 2006) to eject the RAFT-1 (Radar Fence Transponder-1) and MARScom (Military Affiliate Radio System Communication) satellites built by students of the USNA (US Naval Academy), Annapolis, MD.
Since the PSSCT satellite had no active attitude control, it was spun up along its minimum moment of inertia axis prior to ejection from the space shuttle to stabilize its desired orientation: the long sides of the satellite normal to the Earth-sun line.
Figure 4 shows the rate gyro and accelerometer data taken every 0.193 s from 10 to 610 s after spacecraft activation. The acceleration data are averaged over a 1 s interval to reduce the average noise level while the rate gyro data are unprocessed. The spacecraft started with a Z-axis rotation rate of 0º/s while in the launch tube, and reached a rate of 507º/s after 295 s. - Note that the rotation rate does not asymptotically approach the maximum rate and instead it abruptly stabilizes at the 507º/s rate. This is due to an increase in drag torque in the motor as the wheel rotation rate approaches zero. Identical abrupt transition behavior is seen in the rotation-induced radial accelerations measured by the X and Y accelerometers also shown in Figure 4. The difference in final acceleration level between these two acceleration sensors is due to their different distances from the axis of symmetry (0.75 vs. 1.1 cm) and their different bias offsets.
Figure 4: Angular rates and accelerations measured during the “spin-up” phase (image credit: The Aerospace Corporation)
Orbit: Initial orbital parameters for the PSSCT are 51.6º inclination and 345 km altitude. Estimated orbital lifetime is about one year.
Figure 5: Photo of the PSSCT nanosatellite after ejection from the Shuttle Endeavour (image credit: NASA)
Sensor complement: (VGA cameras)
Inexpensive, low power VGA-resolution image sensors are used to provide pictures of the Earth, and to provide extra attitude information. Use of C328 Comedia camera boards; they feature integrated JPEG (Joint Photographic Experts Group) data compression. These boards of size 2 cm x 2.8 cm have a mass of 10 g, a maximum power usage of 200 mW, and contain a 1 cm 640 x 480 color pixel CMOS imager with a lens. The wide field cameras (123º diagonal) are mounted on the +Z and –X sides, while narrow field (56º diagonal) cameras are mounted on the +Z and +X sides.
Mission results and status:
The nanosatellite was spun up along its minimum moment of inertia axis prior to ejection from the space shuttle in order to stabilize its desired orientation: the long sides of the satellite normal to the Earth-sun line. In addition to performing its solar cell testing mission, the satellite provided valuable spin dynamics information. Its temperature-stabilized MEMS rate gyros measured spin rates to better than 1º/s and were sampled at 0.016 Hz for the entire mission. It is known that spinning a spacecraft about the axis of minimum moment of inertia eventually results in a transfer of rotational energy from the minimum axis to the maximum axis or axes, but the time constant was unknown for this satellite design Ref. 1).
The PSSCT nanosatellite went from a ~6º nutation angle to ~60º in about a week. In addition and unexpectedly, the LEO PSSCT also lost momentum rather quickly. The use of a thick aluminum outer shell enhanced eddy-current-induced energy loss rates, applying an external torque. After 77 days on orbit, the LEO PSSCT had gone from a Z-axis rotation rate of 507º/s to less than 5º/s on all axes.
Eddy current damping is the logical explanation for the observed rotational energy loss over time. The eddy current energy loss rates were always an order of magnitude larger than the internal dissipation-induced energy transfer rates from the minimum to maximum axes.
The LEO PSSCT nanosatellite provided data for 109 days until March 18, 2009.
Spin-down - first 9 days:
Figure 6 shows the Z, X, and Y rotation rates as a function of time for the first 213 hours after ejection from the Shuttle. These data were taken using a 64 second sampling period. The Z-axis rate drops from an initial 507º/s to 257º/s after 94.3 hours with an initial Z-axis 1/e decay time of ~140 hours (5.8 days). Over the same period, the X and Y axis rates increase from 19º/s to 107º/s. There are almost 12,000 time samples in Figure 6, so the oscillatory behavior of the X and Y axes, even though they are significantly undersampled, becomes a solid color.
Figure 6: Angular rates measured during the first 213 hours (~ 9 days) of flight (image credit: The Aerospace Corporation)
Spin-down - first 45 days:
Figure 7 shows the Z and Y rotation rates over the first 45 days of flight. The Y and X axes were sinusoidal over this time scale, but that structure is not visible at this image scale; there are 60,800 individual time steps shown with occasional data dropouts (also shown). The X-axis envelope is identical to the Y-axis envelope, so it is not plotted. After 25 days of flight, the second momentum wheel was activated less often and fewer discontinuities in the rotation rate data appear. The Z axis decay rate slows down and the rotation rate at 45 days is 8.2º/s. The Y-axis rates decrease at a much slower pace than the Z-axis. At 45 days, the X and Y axes have peak amplitudes of 20.4º and the LEO PSSCT is primarily in a flat spin.
Figure 7: Z-axis and Y-axis rotation rates for the first 45 days of flight (image credit: The Aerospace Corporation)
Spin-down - beyond 45 days:
The LEO PSSCT provided data for 109 days until March 18, 2009. During the period from 45 to 109 days, the LEO PSSCT was in a flat spin mode and continued to slow down. Figure 8 shows the Z and Y axis rotation rates from 45 to 90 days while Figure 9 shows the X-axis rate. On day 51, the spacecraft began spinning primarily about the Y-axis for 25 days.
The momentum wheel was briefly turned on on day 72 and that reversed the Y-axis rotation rate. X-axis rates generally decreased, with temporary increases after momentum wheel operations. Magnetic torque experiments started on day 73, when rotation rates were low enough to enable real-time torque control. The spacecraft was now subject to operator-induced changes in spacecraft angular momentum. By day 90, all rotation rates were below 5º/s.
Figure 8: Measured Z and Y-axis rotation rates between 45 and 90 days (image credit: The Aerospace Corporation)
Figure 9: Measured X-axis rotation rates between 45 and 90 days (image credit: The Aerospace Corporation)
1) Siegfried W. Janson, David A. Hinkley, “Spin Dynamics of the Pico Satellite Solar Cell Testbed Spacecraft,” Proceedings of the 23nd Annual AIAA/USU Conference on Small Satellites, Logan, UT, USA, Aug. 10-13, 2009, SSC09-IV-5
2) E. J. Simburger, S. Liu, J. Halpine, D. Hinkley, J. R. Srour, D. Rumsey, H. Yoo, “Pico Satellite Solar Cell Testbed (PSSC Testbed),” Proceedings of 2006 IEEE 4th World Conference on Photovoltaic Energy Conversion, Waikoloa, Hawaii, USA, Vol. 2, pp. 1961-1963, May 7-12, 2006., ISBN: 1-4244-0017-1
3) Analog Devices data sheet:, URL: http://www.analog.com/static/imported-files/data_sheets/ADIS16250_16255.pdf
4) Analog Devices data sheet:, URL: http://www.analog.com/static/imported-files/data_sheets/ADIS16003.pdf
5) Siegfried W. Janson, “Micro/Nanotechnology for Picosatellites,” Proceedings of the 22nd Annual AIAA/USU Conference on Small satellites, Logan Utah, USA, Aug. 11-14, 2008, .SSC08-VII-6
The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: "Observation of the Earth and Its Environment: Survey of Missions and Sensors" (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates.