14 January--NASA has opted for copper over optical fiber for connecting components in the next-generation space telescope. In a situation somewhat akin to telecom operators' widespread use of DSL technology instead of fiber optics for broadband service, the James Webb Space Telescope (JWST) will use advanced twisted-pair technology to bridge instruments and electronic components across it's tennis-court-size frame, according to NASA.
NASA engineers say optical-fiber interconnect technology is still not up to the challenges of space. ”Fiber-optics drivers and receivers aren't at a state where the space industry can utilize them,” says Mark Voyton, data-handling systems manager for the JWST. ”However, I expect future generations to be fiber.”
The JWST will use SpaceWire, a European standard for high-speed links and networks that was specifically developed for use on spacecraft. SpaceWire was originally developed by the European Space Agency (ESA), and it has now been improved upon by a team at NASA's Goddard Space Flight Center, in Greenbelt, Md. NASA says its engineers developed a small and very-low-power microchip that can communicate at speeds of more than 200 megabits per second.
Using the SpaceWire protocol in combination with the new transceiver chip, ”we were able to do on one link what would have taken multiple links,” said Voyton. ”We got a 2-to-1 reduction, and 16 wires were reduced to 8.”
The JWST, planned for launch in 2013, will be a large space-based infrared telescope. Viewing the universe in the infrared from the ground is a difficult endeavor, as the water vapor in the Earth's atmosphere absorbs infrared radiation from space. Yet a host of interesting cosmic phenomena--the birth of planetary systems around stars and the birth of stars themselves--are best observed at infrared wavelengths. Astronomers have tried to get around the problem by putting infrared telescopes on dry mountaintops. For example, the Infrared Telescope Facility (IRTF) sits on top of Mauna Kea, in Hawaii. At roughly 4 kilometers above sea level, the IRTF is above most of the water vapor in the atmosphere. Even so, there is still enough air at that elevation to block many portions of the infrared spectrum. A space-based infrared telescope would give astronomers a clearer view into the universe because its mirrors would not be obscured by water vapor.
The JWST is an international collaboration among NASA, the ESA, and the Canadian Space Agency. Named after a former NASA official, the telescope is loaded with new technologies, including a large segmented primary mirror 6.5 meters in diameter, which will be folded during launch and then deployed in orbit. The telescope's optics will be made of ultralight beryllium, and it will have detectors capable of recording extremely weak signals. There will be four scientific instruments on the JWST: a near-infrared (IR) camera, a near-IR multiobject spectrograph, a mid-IR instrument, and a tunable filter imager. The JWST's instruments will work primarily in the infrared but have some capability in the visible range, being sensitive to red, near-, and mid-infrared light at wavelengths from 0.6 to 27 micrometers.
The SpaceWire communications system is driven by low-power chips.
By the time the JWST launches, the current infrared Spitzer Space Telescope will be years past its expected lifetime. SOFIA, an infrared telescope housed in a modified jet, is undergoing flight tests.
The JWST's science instruments will have a total of 66 million detector pixels, and handling the large volume of data they will generate presented a peculiar challenge. But the SpaceWire standard, defined in the European Cooperation for Space Standardization standard ECSS-E50-12A, was up to it. It was authored by Steve Parkes of the University of Dundee, in Scotland, with contributions from the SpaceWire Working Group of the ESA, the European space industry, academia, and NASA.
To achieve faster communication, SpaceWire uses what's called data-strobe encoding, where two signals (data and a strobe) are sent over four wires (two differential copper pairs) in each direction. The data and the strobe signal are configured so that only one of them changes in each clock cycle. Feeding both signals into an exclusive OR (XOR) logic circuit reproduces the line's clock signal.
”Data-strobe encoding allows for higher bandwidths over longer cable lengths, resulting in more flexible system solution options,” says Voyton.
In addition, SpaceWire doesn't need a lot of power, and it is easy to implement a SpaceWire interface in any digital ASIC or FPGA device, Voyton says. Devices including drivers and receivers can be made on a single chip.
SpaceWire helps facilitate the construction of high-performance onboard data-handling systems, reduces system integration costs, increases compatibility between data-handling equipment and subsystems, and encourages reuse of data-handling equipment across several different missions, says Voyton, who has been with the implementation team from the start.
Voyton and colleagues looked at a number of protocols--including the IEEE 1394 FireWire standard and some developed for fiber optics--before settling on SpaceWire.
”We selected SpaceWire based on cost, low power, ease, and flexibility of instrumentation,” Voyton explains.
As a result of being included in the JWST, other missions are also considering SpaceWire. These include NASA's Lunar Reconnaissance Orbiter and its Geostationary Operational Environmental Satellite R (GOES-R). The standard is also being used for technology development at other NASA centers.
”To understand SpaceWire's benefit, compare the speed of a dial-up modem to a high-speed broadband Internet connection,” Voyton said. ”That's what SpaceWire helps us achieve.”