Updated date:

What Were the Gravity A and Gravity B Probes and What Did They Do for Relativity?

Leonard Kelley holds a bachelor's in physics with a minor in mathematics. He loves the academic world and strives to constantly explore it.


The story of these two probes begins with the idea for some new tests of relativity, which came from George Pugh (from MIT) and Leonard Schiff (from Stanford University). In late1959/early 1960, they independently of one another wondered about gyroscope applications in space. And in late 1960 Schiff polished up the details of such a test with some help from William Fairbank and Robert Cannon after advances in gyroscope technology made the envisioned experiment a possibility. 1962 saw Francis Cavoritt join the team, where he would eventually become the principal investigator on Gravity Probe B. Together, the group finally gets funding from NASA in March of 1964 and Gravity Probe A was a go (Kruesi 26, Everitt 5, Ornes).

Gravity Probe A and B

Not much can be said of this first mission simply because not much happened. Launched on June 18, 1976 as a joint venture between NASA and the Smithsonian Astrophysical Observatory, Gravity Probe A’s mission in space lasted 1 hour and 55 minutes as it orbited 6,200 miles above the Earth and then fell into the Atlantic Ocean. The profile called for seeing how gravity affected time, and with an atomic clock that emitted microwaves at a frequency of 1.42 GHz on board scientists were able to compare it with the time from a control clock back on Earth. Results showed that as distance from the Earth increased, time ran at a faster rate, just as Relativity predicted. How much of a change was found, for the sake of knowing? About 4 parts per 10,000 (Kruesi 26, Than).

The calibration telescope.

The calibration telescope.

The cooling tank and housing

The cooling tank and housing

The gyroscope.

The gyroscope.

The Followup...Someday

Amazingly, it took over 40 years for the follow-up mission to Probe A. But why? Many reasons, including 11 management and equipment production challenges. Below are just a few pieces of technology that Probe B developed from scratch (Kruesi 27):

  • High precision gyroscopes
  • Gyroscopic trackers
  • High precision GPS
  • Cryogenic equipment

And the official timeline of Probe B hints at these challenges, for in June 1977 the project status was changed from “exploratory research” to “technology development.” 1982 would see a new study that hinted at high costs which led to a 1983 revision of the project’s goals. Finally, in 1994 (30 years post initial funds), Probe B was given deemed a flight mission with a target launch of October 2000, after over 7 cancellations seen during its lifetime. In the end, 2004 would see the actual lift off because of an unforeseen heating problem in 1998. Turns out, the big cryogenic tank wasn’t able to keep craft cool enough, for 4 windows on the probe needed to let visible light in for tracking purposes but they failed to reflect infrared at the required level. The team had two choices: take the probe apart and replace the windows which would cost two years or drill control pins into the probe which would add 7 months. Option 2 was deemed the best and so they proceeded with that at a careful pace as to not damage any of the components. Finally, after waiting over 40 years, Francis Cavoritt finally say his $750 million mission fly on April 20, 2004 from Vandenberg Air Force Base on a Boeing Delta II rocket, under the leadership of C.W.F. Everitt (Kruesi 27, Ornes).

Mission Objectives

Okay, so I admit I have held out for too long on what all of this builds towards. Gravity Probe B had some predictions of Einstein’s relativity to test including the geodetic effect (GE) and the frame dragging effect (FDE), both results of an object moving through space-time. To be more specific, the GE is movement of an orbiting object causing it to tilt to the side while the FDE is a result of the spinning earth pulling on space-time. To test whether these happen to the levels that relativity predicts, scientists lined up Probe B with IM Pegasi and expected a total shift of 6,606 microseconds a year from the GE and 39 microseconds a year from the FDE. At an orbital height of 399 miles from Earth and orbiting from pole to pole every 97.5minutes, not much can affect such a probe but special conditions were required to have the onboard gyroscope pointed the correct way (Kruesi 26-7, NASA, Ornes).

The goal of the mission.

The goal of the mission.

Gravity Probe B Components

Included in the mission was (Kruesi 26, Everitt 7):

  • A sunshade
  • A telescope to help keep pointing to IM Pegasi (more on that later)
  • Some solar panels
  • 4 gyroscopes
  • 16 micro thrusters which ran on helium
  • A 645-gallon container full of liquid helium (which kept Probe B at -456 Fahrenheit)

Much of this equipment had to have special conditions. For starters, the gyroscopes had to be very smooth and spherical, to an order of no deviations greater than 50 nanometers (in actuality they ended up having only 10 nanometers). They also had to be electrically smooth to an order of 5 parts in 10 million and could not be warmer than 1.95 K (in actuality they were no warmer than 1.8 K). Finally, unaccounted accelerations could be no larger than 1*10­-10 g (actual: less than 5*10-12 g), background magnetic fields could be no larger than 10-6 Gauss (actual: less than 10-7 Gauss), and the pressure inside could be no larger than 10-11 Torr (actual: exact specification). Earth’s surface has 1 g, 0.5 Gauss, and 760 Torr conditions, so hopefully it is clear why Probe B needed the conditions of space to test out this part of relativity (Kruesi 28, Everitt 6-10, Cho, Ornes).



Data Collection and Results...Eventually

Once Probe B was completed with its mission, it had orbited the Earth over 5000 times in a 50-week period from August 27, 2004 until August 14, 2005. It was officially decommissioned in December 2010. Over the first month of the mission, the gyroscopes were slowly rotated inside an electric field until they reached 4000 rotations per minute! To keep the gyros cool, liquid helium was injected into the chamber as they spun, creating a magnetic moment which actually helped us measure the change in the spin direction! Two of the gyros spun clockwise while the other two spun counter clockwise, but all four had their axis of rotation pointed at IM initially. Any deviation the probe felt from external sources were corrected for by thrusters (Kruesi 27-8, Cho).

After all the data had been sent back to Earth, interpreting the GE info wasn’t too hard, with a January 2007 release. The measured effect was 6602 milliseconds ± 33 milliseconds, well within the expected value. The FDE information, however, needed some cleaning up. Turns out, the rotors of such high spherical nature were not perfect in composition and thus caused some voltages to form on the surfaces of the gyroscopes that were unexpected. But fortunately, scientists noticed an integer pattern in the xyz movement of the gyros and were able to parse out the noise. May of 2011, over 5 years after the end of the mission, saw the release of those results, with a FDE measured to be 37.2 milliseconds ± 7.1 milliseconds, also within range of the prediction. But not all are convinced that the data is reliable. At the time of B’s end, indications were that systematic errors would be 10 times larger than were finally reported in 2011. How so much could be eliminated seems to strike some scientists as fishy (Kruesi 28-9, Cho, Than).

And I have hidden the most surprising aspect of all: the FDE didn’t even need B to confirm it. Turns out, evidence for it was found in 2004 when Ignazio Cifuolini and Erricos Pavilis used the LAGEOS and LAGEOS II satellites (both reflector telescopes launched in 1976 and 1992). By measuring deviations in the reflections between the crafts as they orbited the Earth as they precessed about, scientists got to 10% accuracy, better than B’s 19%. This has led to some ridicule of the investment and value of Probe B, but note that over 100 Ph. D. candidates got their doctorates because of this project (Cho).

Works Cited

Cho, Adrian. “At Long Last, Gravity Probe B Satellite Proves Einstein Right.” News.sciencemag.org. AAAS, 04 May 2011. Web. 28 Jun. 2016.

Everitt, C.W.F. et al. “Gravity Probe B Data Analysis – Status and Potential for Improved Accuracy of Scientific Results.” Gravity Probe B Science Results – NASA Final Report Dec. 2008: 5-10. Print.

Kruesi, Liz. “Probing Einstein’s Relativity.” Astronomy Mar. 2012: 26-9. Print.

NASA. “Gravity Probe B Confirms Two Einstein Space-Time Theories.” Astronomy.com. Kalmbach Publishing Co., 05 May 2011. Web. 28 Jun. 2016.

Ornes, Stephen. “Einstein Was Right.” Discover July 2007: 13. Print.

Than, Ker. “Einstein Theories Confirmed by NASA Gravity Probe.” Nationalgeographic.com. National Geographic Society, 05 May 2011. Web. 28 Jun. 2016.

© 2016 Leonard Kelley

Related Articles