Table of contents

A precision comparison between the Josephson series-array voltage standards of the Danish Institute of Fundamental Metrology (DFM) and of the Physikalisch-Technische Bundesanstalt (PTB) has been made at the one volt level by calibrating the same Zener reference standard with both instruments. Within 2.5 nanovolts (1 σ), the same reference voltage has been obtained from the experiments. The calibrated Zener instrument has at the same time been used to measure the average output voltage of the PTB's main group of Weston cells.

A versatile combined primary standard of the USSR has been established at VNIIOFI for the realization of radiometric units. The standard is based on two black bodies, operable at temperatures from 400 to 3000 K, and an absolute radiometer. Units of both spectral and broadband quantities are reproduced within the 0,2 μm to 25 μm range. A black-body radiator made of niobium carbide and graphite is used over the temperature range of 1500-3000 K, while that for the lower temperatures from 400 to 1200 K is a heat-pipe with a sodium heat-transfer medium. In order to determine the thermodynamic temperature T of the high-temperature black body, relative values of radiant flux P_e (T_i) and spectral radiance L_eλ (λ_i, T_i) at several temperatures T_i are found. The method allows realization of radiometric and photometric quantities irrespective of the International Practical Temperature Scale (IPTS-68). The absolute radiometer used is of a special design and has a control system capable of maintaining the conical receiving cavity at constant temperature throughout the measurement process. The absolute radiometer is intended to realize radiometric units as well as to measure the spectral sensitivity of radiation receivers.

The paper concerns the frequency stabilization of an Ar⁺ laser at 514.5 nm to saturated absorption resonances in an external luminescent I₂ cell. With the laser frequency stabilised to the hyperfine component, a₂, a long-term stability Δv/v = 5 × 10^-15 has been achieved for τ = 100 s. The laser linewidth is Δv ≃ 5 kHz. The factors affecting the frequency reproducibility of an Ar⁺ laser have been explored.

In the Physikalisch-Technische Bundesanstalt (PTB), a research test facility for the accurate measurement of gas (volume and mass) flowrates has been set up in the last few years on the basis of a laser Doppler anemometer (LDA) with a view to directly measuring gas flowrates with a relative uncertainty of only 0,1%. To achieve this, it was necessary to develop laser Doppler anemometry into a precision measuring technique and to carry out detailed investigations on stationary low-turbulence nozzle flow. The process-computer controlled test facility covers the flowrate range from 100 to 4000 m³/h (~0,03 - 1,0 m³/s), any flowrate being measured directly, immediately and without staggered arrangement of several flow meters. After the development was completed, several turbine-type gas meters were calibrated and international comparisons carried out. The article surveys the most significant aspects of the work and provides an outlook on future developments with regard to the miniaturization of optical flow and flowrate sensors for industrial applications.

A fixed-point technique for calibrating infrared thermometers between the Zn (420°C) and Cu (1085°C) points was analyzed. To do this, a precision silicon detector thermometer working at 900 nm and 950 nm (band-widths of 70 nm and 40 nm, respectively) and a set of four fixed-point black-body cavities (Zn, Al, Ag and Cu) were used. The best results showed that the ITS-90 can be approximated to within ±0.05°C throughout the Zn-Cu interval when the thermometer in its 900 nm configuration is calibrated at the Zn, Al and Cu points.

Data relating the resistance of high temperature platinum resistance thermometers with thermodynamic temperatures, which were independent from those used to derive ITS-90, have been compared with ITS-90. Within the accuracy of the photoelectric pyrometry involved, ±0,40°C at the freezing point of silver, the measured thermodynamic temperatures agreed with the temperatures obtained using the ITS-90. In measurements performed during a 1,5 year period the typical random uncertainties, evidenced by variations in the resistance of platinum resistance thermometers at the triple point of water, were equivalent to ±0,015°C at the freezing point of silver, whilst, in this period, the drifts in the measured values of the resistance ratio at the freezing point of silver were equivalent to 0,03°C.

``Melting'' and ``freezing'' behaviour of Ga-In eutectic has been studied in order to find the reason for the observed irreproducibility of the ``melting'' and ``freezing'' points of binary eutectics. It is shown that reproducible results can be obtained using techniques of static adiabatic calorimetry and that dynamic methods are inappropriate if one intends to realize liquidus points of eutectics reproducibly with the accuracy necessary for high precision thermometry. The liquidus temperature of Ga-In eutectic was found to be 15.654°C. The eutectic composition is 20.5% In in Ga by weight. This value was obtained by assuming that the heat of fusion per unit mass is maximum at the eutectic composition. The value is 74 J/g.

The change of the apparent mass of 20-g masses and tubes that are not in thermal equilibrium with ambient air has been observed. Buoyancy, adsorption and convection influences are discussed. Quantitative comparisons show that, under such conditions, it is predominantly free convection forces which change the apparent mass.

Argon triple point calibration facilities have been compared among eight laboratories with one transfer system, employing local long-stem standard platinum resistance thermometers. The apparatus intercompared, included a sealed cell and its associated cryostat. As is evidenced by the results of long-term investigations, previously performed at the INM, cells of the type employed may show a triple-point temperature, which is stable within the reproducibility of the measurements (≃0,1 mK) over a period of about 10 years. At each laboratory the mean difference between the Argon triple-point temperature of the transfer cell (t) and the local cell (i) has been determined, using a standard resistance thermometer previously calibrated at the fixed points, according to the IPTS-68; associated repeatabilities are typically of the order of 0,1 mK. The reproducibility attained by measuring the mean difference in different laboratories, using cells of the same type and origin (INM), amounts to 0,4 mK.