Case Closed on the Pioneer Anomaly

Article written: 18 Jul , 2012
Updated: 23 Dec , 2015

Caption: An artist’s view of a Pioneer spacecraft heading into interstellar space. Both Pioneer 10 and 11 are on trajectories that will eventually take them out of our solar system. Image credit: NASA

The case of the Pioneer Anomaly has intrigued and perplexed scientists, engineers and the space-savvy public since 1980, when analysis of tracking data from the twin Pioneer spacecraft showed a small, unexplained slowing of the duo. The answer to this puzzle — now firmly found — lies not in weird physics or mysterious dark matter, but simply the effect of heat pushing back on the spacecraft – heat from the spacecraft itself, emanating from electrical current flowing through instruments and the thermoelectric power supply.

If you’re thinking, “hasn’t this mystery been solved before?” – you’d be right.

Slava Turyshev from the Jet Propulsion Laboratory has laboriously worked on the project since 2004, recovering files from back corners of NASA closets and boxes that were on their way to the trash, converting 1970s punch card data to today’s digital format, and poring over all the data that the spacecraft have beamed back to Earth from billions of miles away.

Along the way, Turyshev has published a couple of papers on his work (here’s one from 2011), and in April of this year, The Planetary Society – who was supporting in part Turyshev’s research – claimed victory that the Pioneer Anomaly was solved.

But now, Turyshev has officially published his findings in the journal Physical Review Letters, and JPL saw fit to put out a press release.

However, over the years other scientists figured out that the culprit might be the heat coming from the spacecraft’s components. In 2001, for example, a scientist named Louis K. Sheffer published a paper, “Conventional Forces can Explain the Anomalous Acceleration of Pioneer 10” and with some good number crunching, determined that “non-isotropic radiation of spacecraft heat” could account for the slowing and “that the entire effect can be explained without the need for new physics.”

Why Sheffer’s paper wasn’t considered more seriously is uncertain, but perhaps at that time the “new physics” idea – that we may have to revise our understanding of gravitational physics — was more intriguing than a mundane effect like heat from the spacecraft’s systems.

But nonetheless, it appears everyone is satisfied with the explanation dutifully resolved by Turyshev and his team of mostly volunteer helpers. And Turyshev’s description of the effect is beautiful in its simplicity:

“The effect is something like when you’re driving a car and the photons from your headlights are pushing you backward,” he said. “It is very subtle.”

Launched in 1972 and 1973 respectively, Pioneer 10 and 11 are still heading on an outward trajectory from our Sun. In the early 1980s, navigators saw a deceleration on the two spacecraft, in the direction back toward the Sun, as the spacecraft were approaching Saturn. They dismissed it as the effect of small amounts of leftover propellant still in the fuel lines. But by 1998, as the spacecraft kept traveling on their journey and were over 13 billion kilometers (8 billion miles) away from the Sun, a group of scientists led by John Anderson of JPL realized there was an actual deceleration of about 300 inches per day squared (0.9 nanometers per second squared). They were the ones who raised the possibility that this could be some new type of physics that contradicted Einstein’s general theory of relativity.

After that, all sorts of theories surfaced, some fairly wacky, some more serious.

In 2004, Turyshev decided to really dig into the matter and started gathering records stored all over the country to analyze the data to see if he could definitively figure out the source of the deceleration. In part, according to JPL, Turyshev and his colleagues were contemplating a deep space physics mission to investigate the anomaly, and he wanted to be sure there was one before asking NASA for a spacecraft.

And so they went searching for Doppler data, telemetry data, and anything they could find about the spacecraft, including picking the brains of navigators who worked with the spacecraft over the years.

They collected more than 43 gigabytes of data, which may not seem like a lot now, but is quite a lot of data for the 1970s. He also managed to save a vintage tape machine that was about to be discarded, so he could play the magnetic tapes. Viktor Toth from Canada, heard about the effort and helped create a program that could read the telemetry tapes and clean up the old data.

They saw that what was happening to Pioneer wasn’t happening to other spacecraft, mostly because of the way the spacecraft were built. For example, the Voyager spacecraft are less sensitive to the effect seen on Pioneer, because its thrusters align it along three axes, whereas the Pioneer spacecraft rely on spinning to stay stable.

Turyshev and his colleagues were able to calculate the heat put out by the electrical subsystems and the decay of plutonium in the Pioneer power sources, which matched the anomalous acceleration seen on both Pioneers.

“The story is finding its conclusion because it turns out that standard physics prevail,” Turyshev said. “While of course it would’ve been exciting to discover a new kind of physics, we did solve a mystery.”

Turyshev’s paper: Finding the Origin of the Pioneer Anomaly.

Source: JPL

, ,

23 Responses

  1. CJSF says

    Wouldn’t it be more accurate to say that for spacecraft like the Voyagers, any possible deceleration isn’t measurable because the use of thrusters to stablize them masks or overwhelms the effect?


    • Olaf2 says

      I think you could say that in the case of Voyager, the emitted heat energy is emitted in a sphere pattern cancelling out the effect. Whereas with Pioneer the loss is in a favourable direction.

  2. lcrowell says

    Years ago and before Perlmutter’s discovery of the accelerated expansion of the universe I ran some calculations to see if the Pioneer anomaly was due to the cosmological constant. The value of the cosmological constant required to induce this was far too large to be physically acceptatble, for the rest of the universe would be vastly different.

    I always figured this was due to some leaking of gas, or the asymmetrical heating of the spacecraft surface or some other basic effect. It is fortunate that somebody worked this long and hard to find what the effect is.

  3. Kawarthajon says

    First, I can’t believe that they were going to throw out all that data/information from the Pioneer missions – isn’t that information of scientific and historic importance? What a great thing that these researchers save all that valuable material! Imagine trying to translate punch cards into digital format – an amazing window on how far our technology has come in such a short time.

    Second – I am a non-science person and I have always been confused about how light can either accelerate things (i.e. solar sail) or decelerate things (i.e. Pioneer) when photons have no mass? Can someone with a science background explain this to me?

    • zetetic elench says

      Yea, what’s up with throwing stuff out over there? Bravo to that fellow’s efforts at explaining this. It’s quite a feat to detect and measure such a discrepancy then go on to explain it. It makes a case for ion propulsion.

    • Olaf2 says

      Photon’s are massless but has energy. When you absorb a photon you receive energy, when you send out a photon (internal heat) then you lose energy. Change in energy means change in orbit.

    • lcrowell says

      You have probably seen this physics already. The Nichols radiometer is that enclosed glass globe with a thin metal vane which rotates on a needle. One side of each vane is colored and the other shiny. Photons are reflected off the shiny side and are absorbed on the colored side. The difference in photon pressure causes the little vane to start rotating around.


      • IVAN3MAN_AT_LARGE says

        Actually, what you describe is a Crookes radiometer, not the Nichols radiometer – which consists of a pair of silvered mirrors suspended by a fine quartz fibre in a torsion balance arrangement within an evacuated glass enclosure.

      • lcrowell says

        I forgot about that distinction. The Crookes radiometer is due to a differential momentum transfer of atoms hitting the vanes.

      • IVAN3MAN_AT_LARGE says

        Yes, and the Crookes radiometer only works if there is a partial vacuum in the glass bulb, but not with a hard vacuum in it.

    • Photons may not have much mass, but gravity does affect them.

    • magnus.nyborg says

      Photons have no rest-mass, but then they also never rest but always travels at the speed of light.
      With classical physics this could not be, but General Relativity allows for massless (restmass) objects to exist if they travel at the speed of light, but only if they travel at the speed of light, and that also allows them to contain energy and momentum.
      The important part here is that photons carry a non-zero momentum, and this momentum can be transferred to other objects causing acceleration or decelleration.

    • Queequeg de la Pequod says

      From ’67 to ’70, I converted Hollerith cards to digital every day at Bell Labs. It worked fine, except when I forgot to remove my correction place marker paper clips from the stack. Today, you should see how fast an envelope (similar sized) flies through a USPS reader.

      • Torbjörn Larsson says

        I hear the story goes that random access memory was invented because you so easily dropped stacks of data on the floor.

      • Queequeg de la Pequod says

        I’m sorry, but I don’t recall that, Senator. Anyway, the “RAM” on our 1ESS telephone switch were actual teensy toroids
        with 3-wires through them. The deck I entered to the ESS had been compiled on an IBM-360 and I had to be able to read and verify the 1s and 0s on the test frame in one of the labs. The permanent memory was on something called a Twister Card. My coding responsibilities were Recent Change Update, Translations and Card Writing (’67-’70, then on to Wisconsin Bell support for AT&T and BTL). BTW, my sister-in-law is 100% pure Lego Little Mermaid Danish.

      • Torbjörn Larsson says

        Yeah, I didn’t mean you personally, General.

        I’m glad your SIL came in Lego, I hear they have awful difficulties with IKEA Swedish Chef blonds in US. I think something about loose screws and incomprehensible instructions.

  4. Zoutsteen from Holland says

    some perspective:
    – diameter of an atom varies between 0.06 to 0.5 nanometers
    > pioneer anomaly …. 0.9 nm/s² deceleration
    – wavelength of visible light is about 380 to 730 nanometers
    – thickness of a human hair around 25000 nanometers

  5. We’ll have to be careful when designing spacecraft then. To have a bit of bleed forward of energy would counteract the thrust of the engines or slow down a spaceship which has reached its cruising speed and is spinning for gravity. On a generations-long trip that could be a factor.

  6. Queequeg de la Pequod says

    And I thought the slowing was due to the solar bow shock. Silly me. Maybe, in 60 years or so, some of our other departing satellites will go down-solar-wind and enter interstellar space toutes de suite.

  7. Torbjörn Larsson says

    Never mind headlights, the pushback is irritating because the way it makes my flashlight jump around. (O.o)

    Good riddance. I remember the crackpots even managed to tie it to timing errors of crafts observed passing close to Earth. Those are all over the place, so it isn’t even a systematic error contribution to pin down but more likely measurement problems. :-/

  8. Gusssss says

    I’ve never seen an acceleration expressed in units of inches per day squared before, and I strongly suspect I never will again.

    At University I received a terse message in red ink when I used the phrase “cycles per second”.

  9. the_only_fan says

    we always make our calculations of gravity towards the sun as the attraction to the sun happens from one point, the center of the sun (we do the same when the earth is involved)
    however as in mechanical constructions there is a difference between a load in one point and a load spread over a length or a surface.
    if we consider the sun not as a point but as a body then a small difference will occur in the gravity results .
    as we take our position as standard there will be a liitle less gravity towards the sun and a little more gravity when we leave our orbit away from the sun.
    In case of the more gravity when we go to the outer space of the sun, it will happen with an decreasing increase in comparison to our position.
    in the case of the “going away from the sun”, it will finally flatten to a constant (the difference in the attraction to the sun will be constant). Towards the sun will be an increasing increase. Try first this to solve that anomaly

Comments are closed.