G using a torsion pendulum time of swing
Dr Jun Luo, HUST Wuhan, China
A new determination of the Newtonian gravitational constant G is presented by using a torsion pendulum with the time-of-swing method. Compared with our previous measurement with the same method, several improvements greatly reduced the uncertainties as follows: (1) measuring the anelasticity of the fiber directly;(2) using spherical source masses minimizes the effects of density inhomogeneity and eccentricities;(3) using a quartz block pendulum simplifies its vibration modes and minimizes the uncertainty of inertial moment; (4) setting the pendulum and source masses both in a vacuum chamber reduces the error of measuring the relative positions. We have performed two independent G measurements, and the two G values differ by only 9 ppm. The combined value of G is 6.67349(18) 10-11m3kg-1s-2 with a relative uncertainty of 26 ppm.
G using a torsion-strip balance
Dr Terry Quinn CBE FRS, BIPM
Two measurements of G have been made at the BIPM, the first published in 2001 and the second in 2013, which agree to within the standard uncertainty of each. Both used the same principles of operation, although for the second measurement the apparatus was almost totally rebuilt, namely, a torsion balance with four masses suspended from a thin wide torsion strip that could be used in two different methods of measurement. One was a simple deflection method, which we refer to as the Cavendish method, and the other using electrostatic servo-control. The advantage of a wide heavily loaded torsion strip is that the restoring torque can be made almost entirely gravitational and thus lossless. The Cavendish and servo methods are largely independent so that if the results agree, as they did, the range of possible systematic errors is constrained to those common to both. Included in these are dimensional metrology and density uniformity in the masses which were carefully studied.
G with a suspended laser interferometer
Dr Harold Parks, Sandia National Laboratories, USA
We will discuss our 2004 measurement of the Newtonian constant of gravitation G using a suspended Fabry-Perot interferometer. The apparatus consists of two simple pendulums hanging from a common support. Each pendulum has a length of 72 cm and their separation is 34 cm. A mirror is embedded in each pendulum bob forming the Fabry-Perot cavity. A laser locked to the cavity measures the change in pendulum separation as the gravitation field is modulated due to the displacement of four 120 kg tungsten masses. We obtain a value of (6.67234 ± 0.00014) × 10-11 m3 kg-1 s-2 for G which is well below the current CODATA value.
Measurement of G using a cryogenic torsion pendulum
Professor Riley Newman, University of California at Irvine, USA
Results of G measurements from years 2000 to 2006 are reported. The G determination is based on measurement of a roughly 1 millisecond change in the 130 second torsional period of a pendulum suspended between a pair of ring-shaped copper masses when the ring pair is rotated 90 degrees. Measurements were made using three different torsion fiber materials (as-drawn CuBe, heat treated CuBe, and as-drawn Al5056), operating at multiple torsional amplitudes. Features of the measurements include: 1) operation at cryogenic temperature near 3K, for reduced anelastic fiber effects, reduced thermal noise, and increased frequency stability, 2) operation at large pendulum oscillation amplitudes corresponding to extrema of the torsional period shift, to reduce effects of torsional amplitude error, 3) the pair of source mass rings, which produced an extremely uniform field gradient, and 4) a thin plate pendulum, minimizing sensitivity to pendulum imperfections. The large distance (about 33 cm) between pendulum and source masses facilitated multipole analysis of their coupling and reduced sensitivity to source mass and pendulum imperfections. A price paid for this was the extremely small period change signal, ranging from 1.7 to 0.2 milliseconds depending on fiber and torsional amplitude