225 Years Later, We Still Can't Measure Gravity — The Big G Problem

Every other fundamental constant in physics has been pinned down. The speed of light is fixed by definition. Planck's constant is exact. The fine-structure constant is known to better than a part in a billion. Then there is big G — the number that sets the strength of gravity itself. Newton wrote it into his law. Einstein put it at the centre of his. And 225 years after Henry Cavendish first measured it with lead, wood and wire in a sealed shed, the best laboratories on Earth still cannot agree on its fourth digit. In April 2026, NIST published a new measurement using the same torsion balance the French had built decades earlier. Same design. Same geometry. Different continent. The answer came back about 250 parts per million lower. This is the story of the least well known number in physics: why gravity is so absurdly weak that measuring it is almost impossible, how a painfully shy millionaire "weighed the world" in 1798, why a dozen state-of-the-art experiments disagree by more than their own error bars allow — and why the mass of the planet you are sitting on is only as precise as big G. CHAPTERS 0:00 Two opposite mysteries 0:54 Newton's law is missing a number 1:25 Why gravity is absurdly weak 2:13 Henry Cavendish and John Michell 2:55 Inside the torsion balance 3:38 1798: weighing the world 4:02 Every other constant got sharper 4:24 The experiments that disagree 5:15 NIST reruns the French machine 6:02 The sealed envelope 7:07 April 2026: the new value 7:32 So what is going wrong? 8:18 Earth's mass is only as good as G 9:08 1/137 vs big G WHAT YOU'LL LEARN Why Newton's law of universal gravitation needs one extra number, and why Newton could never measure it Just how weak gravity is: between a proton and an electron, the electric force beats it by a factor of about 10^39 How Cavendish's torsion balance landed within 1% of today's value using lead, wood and wire Why about a dozen high-precision measurements of G scatter further apart than their own error bars allow How NIST ran a blind analysis with a secret number sealed in an envelope for two years Why Earth's mass and the Sun's mass are only known to a few digits THE NUMBERS CODATA recommended value: G = 6.67430 x 10^-11 m^3 kg^-1 s^-2 NIST, April 2026: 6.67387 x 10^-11 BIPM, France: about 6.67554 x 10^-11 Atom interferometry, Florence: about 6.67191 x 10^-11 SOURCES AND FURTHER READING NIST — "NIST Weighs In on the Mystery of the Gravitational Constant" (April 2026) Physics (APS) — "Gravitational Constant's Value Still Up in the Air" Cavendish, H. (1798) — "Experiments to Determine the Density of the Earth" WATCH NEXT The Mystery of 1/137: The Number That Haunted Richard Feynman The Mystery of 1/137: The Number That Haun... The Muon g-2 Anomaly: The Particle That Breaks Physics The Muon g-2 Anomaly: The Particle That Br... Your GPS Would Be 10 km Wrong Every Day Without Einstein Your GPS Would Be 10 km Wrong Every Day Wi... If you learned something new, subscribe — it genuinely helps. And tell me in the comments which number you want next. #Physics #Gravity #BigG #GravitationalConstant #Cavendish #Science #Newton #Einstein