<<Pioneer 10 and 11 were sent on missions to Jupiter and Jupiter/Saturn respectively. Both spacecraft were spin-stabilised in order to keep their high-gain antennas pointed towards Earth using gyroscopic forces.
Although the spacecraft included thrusters, after the planetary encounters they were used only for semiannual conical scanning maneuvers to track Earth in its orbit, leaving them on a long "cruise" phase through the outer Solar System.
Because the spacecraft were flying with almost no additional stabilization thrusts during their "cruise", it is possible to characterize the density of the solar medium by its effect on the spacecraft's motion. In the outer Solar System this effect would be easily calculable, based on ground-based measurements of the deep space environment. When all known forces acting on the spacecraft were taken into consideration, a very small but unexplained force remained. It appeared to cause an approximately constant sunward acceleration of (8.74±1.33)×10−10 m/s2 for both spacecraft. If the positions of the spacecraft were predicted one year in advance based on measured velocity and known forces (mostly gravity), they were actually found to be some 400 km closer to the sun at the end of the year. This anomaly is now believed to be accounted for by thermal recoil forces.
Starting in 1998, there were suggestions that the thermal recoil force was under-estimated, and perhaps could account for the entire anomaly. However, accurately accounting for thermal forces was hard, because it needed telemetry records of the spacecraft temperatures and a detailed thermal model, neither of which was available at the time. Furthermore, all thermal models predicted a decrease in the effect with time, which did not appear in the initial analysis.
One by one these objections were addressed. Many of the old telemetry records were found, and converted to modern formats. This gave power consumption figures and some temperatures for parts of the spacecraft. Several groups built detailed thermal models, which could be checked against the known temperatures and powers, and allowed a quantitative calculation of the recoil force. The longer span of navigational records showed the acceleration was in fact decreasing.
In July 2012, Slava Turyshev et al. published a paper in Physical Review Letters that explained the anomaly. The work explored the effect of the thermal recoil force on Pioneer 10, and concluded that "once the thermal recoil force is properly accounted for, no anomalous acceleration remains."
The Pioneers were uniquely suited to discover the effect because they have been flying for long periods of time without additional course corrections. Most deep-space probes launched after the Pioneers either stopped at one of the planets, or used thrusting throughout their mission. The Voyagers flew a mission profile similar to the Pioneers, but were not spin stabilized.
Instead, they required frequent firings of their thrusters for attitude control to stay aligned with Earth. Spacecraft like the Voyagers acquire small and unpredictable changes in speed as a side effect of the frequent attitude control firings. This 'noise' makes it impractical to measure small accelerations such as the Pioneer effect; accelerations as large as 10−9 m/s2 would be undetectable.
Newer spacecraft have used spin stabilization for some or all of their mission, including both Galileo and Ulysses. These spacecraft indicate a similar effect, although for various reasons (such as their relative proximity to the Sun) firm conclusions cannot be drawn from these sources. The Cassini mission has reaction wheels as well as thrusters for attitude control, and during cruise could rely for long periods on the reaction wheels alone, thus enabling precision measurements. It also had radioisotope thermoelectric generators (RTGs) mounted close to the spacecraft body, radiating kilowatts of heat in hard-to-predict directions. After Cassini arrived at Saturn, it shed a large fraction of its mass from the fuel used in the insertion burn and the release of the Huygens probe. This increases the acceleration caused by the radiation forces because they are acting on less mass. This change in acceleration allows the radiation forces to be measured independently of any gravitational acceleration. Comparing cruise and Saturn-orbit results shows that for Cassini, almost all the unmodelled acceleration was due to radiation forces, with only a small residual acceleration, much smaller than the Pioneer acceleration, and with opposite sign.
Before the thermal recoil explanation became accepted, other proposed explanations fell into two classes — "mundane causes" or "new physics". Mundane causes include conventional effects that were overlooked or mis-modeled in the initial analysis, such as measurement error, thrust from gas leakage, or uneven heat radiation. The "new physics" explanations proposed revision of our understanding of gravitational physics.
If the Pioneer anomaly had been a gravitational effect due to some long-range modifications of the known laws of gravity, it did not affect the orbital motions of the major natural bodies in the same way (in particular those moving in the regions in which the Pioneer anomaly manifested itself in its presently known form). Hence a gravitational explanation would need to violate the equivalence principle, which states that all objects are affected the same way by gravity. It was therefore argued that increasingly accurate measurements and modelling of the motions of the outer planets and their satellites undermined the possibility that the Pioneer anomaly is a phenomenon of gravitational origin. However, others believed that our knowledge of the motions of the outer planets and dwarf planet Pluto was still insufficient to disprove the gravitational nature of the Pioneer anomaly. The same authors ruled out the existence of a gravitational Pioneer-type extra-acceleration in the outskirts of the Solar System by using a sample of Trans-Neptunian objects.
Gravitationally bound objects such as the Solar System, or even the Milky Way, are not supposed to partake of the expansion of the universe—this is known both from conventional theory and by direct measurement. This does not necessarily interfere with paths new physics can take with drag effects from planetary secular accelerations of possible cosmological origin.
It is possible that deceleration is caused by gravitational forces from unidentified sources such as the Kuiper belt or dark matter. However, this acceleration does not show up in the orbits of the outer planets, so any generic gravitational answer would need to violate the equivalence principle. Likewise, the anomaly does not appear in the orbits of Neptune's moons, challenging the possibility that the Pioneer anomaly may be an unconventional gravitational phenomenon based on range from the Sun.
The cause could be drag from the interplanetary medium, including dust, solar wind and cosmic rays. However, the measured densities are too small to cause the effect.>>