Another close look at “big G”

In 1798, more than 70 years after Isaac Newton died, Henry Cavendish was the first to measure Newton’s gravitational constant G. He used a torsion balance in which test masses on a suspended balance beam were slightly deflected by nearby source masses. In 2001 physicists at the International Bureau of Weights and Measures (BIPM) in Sèvres, France, used an updated torsion balance to measure G in two substantially independent ways—the standard Cavendish method that relies on measuring the angular deflections and an electrostatic force compensation method that holds the test masses steady against deflections. The two results agreed within their uncertainties but were notably higher than the previously best-determined value. The BIPM team has now rebuilt its torsion balance (seen in the figure) to reduce uncertainties and has redone the measurements. The two methods again generated values that agree with each other; the new combined result is G = 6.675 45(18) × 10⁻¹¹ m³ kg⁻¹ s⁻² with an uncertainty of 27 parts per million. The value statistically agrees with the team’s earlier result but is 241 ppm above the 2010 value recommended by the international Committee on Data for Science and Technology (CODATA). Two of the researchers are organizing a 2014 conference that will feature all experimenters with recent big G results, trying to pin down this poorly known fundamental quantity. (T. Quinn et al., Phys. Rev. Lett. 111, 101102, 2013.)