Jennifer Lauren Lee / NIST PML

Scientists from NIST have created the fourth generation of Watt scales, which are planned to be used to refine the Planck constant and, ultimately, to redefine the kilogram. The new device differs from its predecessor NIST-3 by replacing the expensive to use superconducting magnet with a constant, smaller size and weight, as well as many changes in the design, which have reduced the influence of parasitic effects. The results of the first NIST experiment on a new device are published in the journal *Review of Scientific Instruments* .

In 2014, at the XXV General Conference on Measures and Weights, scientists made the final decision to redefine the unit of measurement of the International System of Units (SI) kilogram using universal physical constants. Until now, the definition is based on a comparison with a reference sample and sounds like: “a kilogram is a unit of measure of mass equal to the mass of an international prototype of a kilogram”. But a platinum-iridium ingot used as a reference can change its mass with time due to evaporation and diffusion. Therefore, the definition of this unit of measurement using physical constants is more universal, and will not depend on the time or storage conditions of the standard.

Currently, many scientific organizations around the world are involved in the redefinition of a kilogram, including the National Institute of Standards and Technology (NIST). It is assumed that the result of their work will be a more accurate value of the Planck constant , which in SI is measured in m ^{2} · kg · s ^{−1} . Fixing its value, a new definition of a kilogram will be introduced, and the mass of the platinum-iridium standard will become equal to it with some error.

NIST experiments are based on the so-called basic watt balance equation, which is the equality of two relations: 1) on the one hand, the ratio of Planck’s constants, one of which is a value in SI, and the other its value in arbitrary units , adopted in 1990 : 2) on the other hand, the ratio of mechanical power in SI to electrical also in arbitrary units of 1990. Thus, to determine the Planck constant in the watt balance experiment, it is necessary to balance the electrical and mechanical force. The values of the 1990 constants were fixed to avoid confusion in the accuracy of the physical constant values used by different groups.

In conventional scales, the mass of the load is compared with a certain standard, for example, with weights of known mass. In the Watt scale, it is not the load that is compared with the standard, but the repulsive force between the permanent magnet and the coil through which current is passed. The basis of the design of the watt scales in NIST is a flywheel, on one side of which the scales are located, and on the other, the engine, which raises the coil at a constant speed in one of the modes.

Between themselves, both parts and the wheel are connected through a complex system of cables. In the “weight” part, the cables support the pallet for the test standard – one “bowl” – and the coil rigidly fixed with it – the other “bowl”. The coil, in turn, is placed in a magnetic field, which is created using a system of two disks of a ferromagnet Sm _{2} Co _{17} with a total mass of about 800 kilograms.

In the “speed” mode, the cargo tray remains empty, and the counterweight engine lifts the coil at a constant speed relative to the magnets. This creates a current of a certain magnitude. Thus, by measuring the voltage in the coil, you can determine the strength of the field created by the magnets.

In the “weighing” mode, the load is placed on the platform, and a current is applied to the coil, which, in turn, creates a magnetic field that interacts with the field of permanent magnets. A pushing force occurs, which ultimately must balance the entire coil-load system between two magnets. Determining what kind of current for this need to apply to the coil, you can determine the magnitude of this force. Comparing it with the force of gravity acting on the weight of a known mass, scientists determined the Planck constant by the watt-balance equation.

The first experiment conducted on NIST-4 allowed us to obtain the value of the Planck constant h = 6.626 069 83 (22) × 10 ^{−34} J ∙ with an accuracy of 34 billionths of a share, while scientists expected the result from the first start to be six times worse. The precision record – 19 billionths of a share – is currently owned by the National Research Council in Canada (NRC), but scientists from NIST plan to reach and overcome this limit with the help of a new device next year.

It is planned that the final experiment to redefine the kilogram will be conducted in 2018. According to the standards, for this it is necessary that the experiment be reproduced in at least three independent laboratories, and at least in one of them the accuracy of the value of the Planck constant was less than 20 billionths of a billion. In addition, at least one experiment should be built on different physical principles.