Automatic weighing plays an important part in many fields of applications: Large quantities of commodities and raw materials are handled in trading centres using automatic belt weighers or totalizing hopper weighers for bulkto- bulk weighing, smaller quantities of commodities being meant for end users are automatically filled and weighed by gravimetric filling instruments, by catchweighers or checkweighers. Commodities transported on vehicles or by rail are often weighed automatically by in-motion road vehicle weighing instruments or automatic railweighbridges, respectively. New developments and measurement techniques are especially observed in the field of waste disposal, weighing of container wagons in cross-border traffic, shovel dozers for weighing building materials, and automation of processes in the food and non-food industry. This paper is intended to give a survey on the state of the art in automatic weighing and on international efforts (eg. OIML) towards harmonization of specifications and test procedures for automatic weighing instruments.

The lever arm length, i.e. the distance between the force-acting line and the rotation axis of the lever, directly defines the torque that is acting in a torque measuring machine of the lever type. Therefore, already in the design process of the 20 kN·m static torque standard machine finite element calculations had been carried out at PTB to optimize the lever geometry in order to guarantee the constancy (within the limits of the uncertainty of measurement of the machine) of the lever arm length, even during loading. The absolute measurement of the lever arm length had still been a problem at this time.
The present work describes the method of defining the lever arm length using a calibrated precision end gauge. With this gauge the above mentioned constancy of the length during the loading procedure can also be proved. Furthermore, the chosen lever geometry has an additional advantage – it permits the very sensitive adjustment of the lever arm length in the µm-range, when the bent side plates are tightened against each other. The measurement results demonstrate the success of this procedure.

A new experiment has been started to determine the atomic mass of gold in the unit kilogram. For this purpose, gold ions have to be accumulated to a weighable mass and the ion current has to be measured and integrated over the accumulation time. In a first accumulation experiment gold ions have been accumulated on a gold coated quartz crystal. A gold ion beam with a current of 5 mA and a range of ion energy between 25 eV and 300 eV has been used. As a result of the interaction of the gold ions with the surface of a gold coated quartz crystal, the ratio of the change of mass per time and the ion current multiplied with the elementary charge has been determined as a function of the ion energy. The transition between the mass loss and the mass increase at the oscillator quartz, mainly resulting from the sputtering effects, has been observed. The measurements have shown a reasonable agreement with the expected value of the mass of a gold atom in the unit kilogram for low ion energies.

Top pan electronic comparators, which are widely used in National Measurement Institutes and mass calibration laboratories, for large mass calibration, inherently suffer from eccentricity and hysteresis. NPL has taken a commercial 60 kg comparator and designed an automatic weight exchanging mechanism to overcome these problems. This has been achieved by using a self centring scalepan with two weighing stations and exchanging between these two stations whilst maintaining a constant load on the comparator.
This paper outlines the initial assessment of the commercial comparator, describes the conceptual design of the weight exchanger, and comments on the improved performance observed from the comparator with the exchanger. The paper also discusses the work currently in progress on a second generation weight exchanger, scaling the project up to 500 kg, incorporating modifications to overcome the problems found in the 50 kg weight exchanger.

To establish a national mass standard, submultiple and multiple systems are employed to calibrate highly accurate sets of weights based on national prototype kilograms. However, calibration techniques require great skill and many years for the measurement of weights. To promote an efficient calibration process, an automatic calibration device with built-in standard weights is being developed. In the first step, a device was manufactured on an experimental basis; this device automatically calibrates a set of built-in ring-shaped standard weights (100 g ~ 1 kg) using the submultiple method. Then, preliminary experiments were performed. In the second step, a weight-exchange mechanism which directly calibrates test standard weights (1000, 500, 200, 200, 100 g) using the submultiple method was added to the device and calibrations were carried out. Results of preliminary measurements reveal a standard deviation of approximately 0.04 ~ 0.02 mg.