Postcards from the edge

By G.F. | PASADENA, CALIFORNIA

VOYAGER 1 has been beaming data to Earth since 1977. But members of mission control at NASA's Jet Propulsion Laboratory (JPL), in Pasadena, California, are as excited as ever. Some time before 2015 the probe should report that it has entered the heliopause, an area where the sun's "solar wind" is no longer strong enough to beat back the stellar winds of neighbouring stars. At that point its "triaxial fluxgate magnetometer" will detect a change in the direction of the magnetic field perpendicular to its path from east-west to north-south. Voyager 1 would then, the American space agency's press office insists, become the first manmade object to leave the solar system.

Cynics—as well as most astronomers, cosmologists and, indeed NASA itself—point out that the edge of the solar system is more properly defined as the point beyond which an object will not succumb to the sun's gravity. Gravity is, after all, what defines the universe at the grandest scale. That, though, is roughly 50,000 times farther from the sun than Earth is. Voyager 1 has so far travelled 123 times Earth's distance from the sun, or 18 billion km. It would need another 14,000 years, give or take, to escape the sun's gravitational pull at its current speed.

None of which should detract from the Voyager programme's momentous achievements. Voyager 1 and its twin, Voyager 2, dispatched 15 days earlier but now trailing because of its excursion to Uranus and Neptune, have offered glimpses of the four gas giants and a slew of strange astronomical phenomena. And while Voyager 1 will remain in the solar system for some time to come, it will soon enter the region where the solar wind's charged particles give way to a thinner medium of dust and other matter that fill space between stars. Scientists keenly await news from that uncharted territory.

Over the years the Voyagers have sprung a number of astronomical surprises. The latest came last summer, when Voyager 1 encountered a previously unknown phenomenon now dubbed the "magnetic highway". In this region, its instruments suggest, solar and interstellar magnetic fields meet. Edward Stone, the Voyager programme's boss since its inception in 1972, explains that this causes them to align in such a way that lower-energy particles from within the "heliosphere" of the sun's magnetic influence seep out, replaced by higher-energy particles from the expanses of space.

Voyager 1 should also soon be able to settle once and for all that as the sun sweeps through the interstellar medium it does not leave behind a wake. The existence of this "bow shock" used to be astronomical received wisdom but evidence collected in the last few years by the Voyagers and the Interstellar Boundary Explorer (IBEX) craft orbiting Earth indicates that the sun must be travelling too slowly relative to the interstellar medium to produce this effect.

To get where they are, both Voyagers first passed through the "termination shock" (Voyager 1 did so in December 2004; Voyager 2 followed suit in August 2007), a point at which the sun's solar wind abruptly slows due to pressure from interstellar gases, and entered the heliosheath. The heliosheath was previously thought to be turbulent, yet the Voyagers have measured no solar wind at all, suggesting that solar and interstellar gales cancel each other out. From there, Voyager 1 has gone on to cruise the magnetic highway, with Voyager 2 expected to do so in the coming years.

The hope was that the probes would be sturdy and resilient enough to withstand the predicted trip across the termination shock and buffeting in the heliosheath, as well as other cosmic vagaries on their journeys, which took both perilously close to Jupiter and Saturn, with Voyager 2 making a detour past Uranus and Neptune. So when in 1973 Pioneer 10, launched a year earlier, measured radiation around Jupiter to be substantially higher than anticipated, Dr Stone's team spent nine months replacing and even redesigning many of the components of the Voyagers, already well in the works in the early 1970s. Of course, the probes were engineered with plenty of redundancy. For example, each carries two copies of three separate computer systems. So far, though, few of the back-up systems ever needed to be deployed. Dr Stone, a glint of paternal pride in his eye, boasts that both vehicles have far exceeded expectations.

Cleverness back on Earth, too, has played a role in the mission's success. When Voyager 2's primary and secondary receivers failed a year into its mission, the ground team worked around it to communicate with the back-up system, which remains in use to this day. In 2010, on receiving a garbled message from the probe, the team painstakingly retrieved a full dump of memory using one of the back-ups, which revealed a single bit in a program sequence had flipped from 0 to 1. Reloading the program did the trick.

Nor is the tinkering confined to fixing glitches. The boffins regularly updated control systems to ensure optimal use of the probes' resources during their most active phases. They did so 18 times during the Jupiter stage of Voyager 1's mission alone. Take data transfer. When Voyager 1 and Voyager 2 flew around Jupiter and Saturn, the probes were close enough to Earth to send uncompressed images and other data at relatively high bit rates of 115,000 and 45,000 bits per second (bps), respectively. Because signal strength varies as the inverse square of the distance between receiver and transmitter, however, by the time it reached Uranus, Voyager 2 would be transmitting at 9,000bps. This would drop to 3,000bps around Neptune, substantially reducing the number of images and readings that could be beamed home.

Most back-up computers are designed to operate only if the primary system fails. However, one of the probes' auxilliary systems was hooked up so it and its primary could run in parallel when needed. Before Voyager 2 flew by Uranus, simple image-compression software was transmitted to the back-up, allowing a 640 kilobyte picture to be squeezed without loss to as little as 256 kilobytes.

Most ingeniously of all, Dr Stone's team equipped the probes with an advanced bit of hardware called a Reed-Solomon encoder. The device significantly reduced the burden of "error-correction" necessary to ensure that a message can still be read if some of the bits are lost or corrupted during transmission. Originally, the Voyagers used an older, well-tested system that sent one "error-correcting" bit for every bit of the actual message; error correction therefore hogged half the available bandwidth. The Reed-Solomon encoder would enable a single bit to correct for five bits of the message. The rub was that in 1977 a way to decode Reed-Solomon corrected data had yet to be worked out. Luckily, by the time Voyager 2 reached Uranus in 1986, it had been.

At present the data trickle in from the Voyagers to the Deep Space Network of radio telescopes sprinkled around the globe (which capture information from all manner of space and planetary operations), and thence to JPL, at just 160bps. This was a conscious decision made years ago to maintain a constant rate for the remainder of the mission. The crafts' cameras were turned off after planetary fly-bys, and just a few instruments remain active. For about 30 minutes every six months data stored during that period on an 8-track digital tape are transferred at a zippier 1,400bps.

The Voyagers' radioisotope thermoelectric generators, ticking away with the decay of their plutonium-238 fuel, should keep the instruments and transponders going until at least 2021. Then, around 2025, after nearly half a century of boldly going where no manmade object had gone before, the team will shut down the probes while communications are still available in a somewhat sentimental effort to keep the Voyagers aimed straight and true. And they will drift deeper into interstellar darkness.