Table of contents

Gas mass flows can be measured most accurately using continuously measuring dynamic-gravimetric systems. Two systems of this type, the mass flow calibration system at the Bureau National de Métrologie-Laboratoire National d'Essais (BNM-LNE, France) and the mass flow calibration system at the Physikalisch-Technische Bundesanstalt (PTB, Germany), were compared twice using a commercial gas flow measuring instrument as a transfer standard. Mass flow rates of nitrogen between 0.5 mg/s and 20 mg/s (approximately 25 ml/min to 1000 ml/min) and mass flow rates of carbon monoxide between 0.5 mg/s and 10 mg/s (approximately 25 ml/min to 500 ml/min) were used to check the agreement of the two calibration systems. The results of the comparison show that the two systems do not deviate by more than 0.23 % for flow rates higher than 2 mg/s. For lower flows, down to 0.5 mg/s, the differences increased to as much as 0.37 %. The reason for this is not yet understood.

The melting temperatures of three metal-carbon eutectics, rhodium-carbon, platinum-carbon and ruthenium-carbon, were measured by four national metrology institutes: the National Research Laboratory of Metrology (NRLM, Japan, now the National Metrology Institute of Japan, NMIJ), the National Institute of Metrology (NIM, China), the National Measurement Laboratory (NML, Australia), and the Korea Research Institute of Standards and Science (KRISS, Rep. of Korea). The same cells were measured using different radiation thermometers and different high-temperature carbon-tube furnaces by the four institutes, on their independently realized temperature scales. The results were compared to examine the performance of these high-temperature fixed points and to determine their melting and freezing temperatures on the International Temperature Scale of 1990 (ITS-90). The measured fixed-point temperatures were all within 0.4 °C of the calculated mean for each cell, and were within each laboratory's estimated uncertainties. The mean values were (1656.8 ± 0.6) °C, (1737.4 ± 0.3) °C and (1953.0 ± 0.4) °C, respectively, for the melting plateaus of the Rh-C, Pt-C and Ru-C eutectic points, with the uncertainties given for k = 2.

The melting and freezing plateaus of three metal-carbon eutectic alloys, cobalt-carbon, iridium-carbon and rhenium-carbon, were observed by radiation thermometry for the first time, and their repeatability evaluated. The observed melting temperatures were 1324 °C, 2290 °C and 2474 °C, respectively. A small dependence of the freezing temperature on the cooling rate was observed for the cobalt-carbon eutectic. Six metal-carbon eutectic points have previously been reported by the present authors: these, with their melting temperatures, are iron-carbon (1153 °C), nickel-carbon (1329 °C), palladium-carbon (1492 °C), rhodium-carbon (1657 °C), platinum-carbon (1738 °C), and ruthenium-carbon (1953 °C). These nine fixed-point temperatures cover the range from the copper point to 2500 °C. The evaluated melting-point repeatabilities (standard deviations) were generally better than 0.1 °C. The metal-carbon eutectic points seem to be a promising way of realizing fixed points for practical calibration of radiation thermometers. The two fixed points above 2500 K in particular are potentially useful for thermodynamic temperature determination based on thermal radiation, as well as for radiometry and photometry.

An evaluation of the light shift in an optically pumped caesium-beam frequency standard is presented. We show that the light shift can be accurately estimated by taking into account the angular anisotropy of the spontaneous photon emission, the neglect of which is shown to give rise to an error comparable to the value of the shift. Our data refer to a complete set of optical transitions for the caesium D₂ line relevant to the KRISS-1 caesium-beam frequency standard.

Six silver samples were doped with Pd, Fe, Au, In, Sn, Pb, Zn, Sb, Cu, Ni and Al. The resulting shifts in the equilibrium melting curves were then determined. All impurities caused a shift in the liquidus point of silver except aluminium, for which no conclusive results could be obtained. The shifts, in millikelvins, caused by one part per million by weight of each of the above impurities (with the exception of the last mentioned, Al), were determined to be +0.97, +0.13, +0.09, −0.49, −0.60, −0.48, −1.10, −0.73, −0.69, and −0.74, respectively. Four of the initial samples were recently acquired and of 99.9999 % purity. Their melting points were realized within a scatter band of ±0.55 mK. Using the above measured shifts and the amount of impurities detected in these samples by glow discharge mass spectrometry (GDMS), it is estimated that the liquidus points of these samples differ from that of ideally pure silver by no more than 1.0 mK within an expanded experimental standard deviation of the mean of 0.2 mK, using a coverage factor of 2.

The most commonly used method in the calibration of step gauges is the coordinate measuring machine (CMM), equipped with a laser interferometer for the highest accuracy. This paper describes a modification to a length-bar measuring machine for the calibration of step gauges to a high accuracy. A system was also developed for interferometric measurements of the flatness and parallelism of gauge block faces for use in uncertainty calculations.

The results are presented of calculation of the frequency dependence of an octofilar resistor, the resistive element of which is folded in such a way that it forms four long loops connected in series. Magnetic fluxes produced by the loops tend to cancel, similar to the fluxes produced by the two loops of the common quadrifilar reversed resistor. The folded length of the octofilar resistive element is one half of that of a quadrifilar resistor made from the same wire and having the same nominal resistance, and the octofilar resistor has a smaller frequency dependence. As an example, a 12906 Ω octofilar resistor is described, for which the frequency dependence of its parallel equivalent resistance is more than seven times smaller than that of a previously made 12906 Ω quadrifilar resistor in the frequency range 500 Hz to 5 kHz.

The upper limit to the power at which trap detectors have a linear response was recently measured. We have complemented these measurements by testing traps with an applied bias voltage at several visible wavelengths. We used Hamamatsu S1337-1010 silicon photodiodes, and also briefly tested the S5227-1010 photodiode. Application of a bias voltage increases the linearity limit by a factor of more than 10 for very narrow beams and of more than 30 for wide beams. No irreversible changes were detected even for the highest irradiance measured (33 W/cm² at 406 nm). The minimum bias voltages required for linearities of 99 %, 99.8 % and 99.95 % are presented for several beam sizes.

The original theory of the Senarmont compensator allows the calibration of a phase plate, C₁, using the optical arrangement P-C₁-C₂-A, where P and A are respectively the polarizer and analyser and C₂ is an exact quarter-wave phase plate. Recently, it was shown that two plates, C₁ and C₂, may be simultaneously calibrated using the same arrangement but with some restrictions on their retardance values at the working wavelengths. A further development of this model is presented that removes the restrictions on the ranges of retardance value allowed for the two plates.

Comparison measurements of the diameter of small ring gauges (0.1 mm to 1 mm) were carried out by six European national metrology institutes (NMIs). The participants applied different measurement principles: mechanical contacting measurement and optical measurement with respect to the projection of the inner diameter or with respect to the position of the cylinder edges. The uncertainties associated with the diameter measurement results vary considerably among the different participants, from U = 10 µm to U = 5 µm. In general the measurement results agree within their associated uncertainties, with the exception of the optical measurements of the position of the cylinder edges which exceed the reference values by between 10 µm and 30 µm, probably because of the lacerated edges of the ring gauges.

This letter takes a brief look at gauge-block length as it is influenced to a subtle degree by the surface quality of the measuring face and the recent history of the gauge block environment. Change in length of short gauge blocks following wringing, and of long gauge blocks following air travel between laboratories, is demonstrated. These effects could have important implications for international comparison testing under time-limited conditions, particularly with the large number of comparisons related to the Mutual Recognition Arrangement (MRA).