Table of contents

The realisation of the normal boiling point of neon is an essential step in implementing the International Practical Temperature Scale of 1968. A cryostat designed for this purpose is described, and the results given of detailed investigations into the stability and uniformity of temperature achieved. Preliminary results are presented of an investigation of the absolute accuracy of the a.c. technique used for resistance thermometry. It is concluded that thermal conditions in the cryostat should permit a repeatability of ± 10^-4 K, while the resistance measurement will not introduce an absolute error greater than 2 × 10^-4 K.

A krypton 86 source, used in the recommended way for the emission of the primary standard radiation, provides also many other radiations, particularly in the near infrared region (λ = 0.7 to 1.2 μm). Among these, nine radiations have been selected; they allow to observe interferences up to path differences exceeding 1 m, with an accuracy to be compared with the one afforded by the primary standard at a path difference of 0.6 m. Their wavelengths have been measured with an accuracy of about 3 parts in 10⁹ and the broadening and shift effects due to the bath temperature and the current density in the discharge have been investigated. Only one radiation exhibits a temperature-shift definitely larger than the one exhibited by the primary standard.

The NPL Lumen [2793 K] unit is preserved in the calibration of a group of gas-filled, tungsten filament, flux standards operating at a colour temperature of 2793 K. Its value relative to the NPL Candela [2045 K] has been established with an accuracy of ± 0.2% by means of an automatic recording goniophotometer. – A subsidiary experiment on high-pressure mercury vapour lamps showed that flux measurements using an integrating sphere method are in good agreement with those found using the goniophotometer.

The reproducibility of the IPTS-68 as determined by forty-eight thermometers is shown to be within approximately ± 3 mK in the range 13.81 K to 20.28 K, ± 1 mK in the range 20.28 K to 54.361 K, and ± 2 mK in the range 54.361 K to 90.188 K. In the range 90.188 K to 273.15 K, nine thermometers exhibit a scale reproducibility within only ± 5 mK; it is believed that this figure is limited by calibration uncertainties, and that ± 1 mK is more likely to be typical.

A simple passive cavity CO₂ laser has been designed and shown to be ideally suited for length measurement applications, particularly for the calibration of long tapes. The laser wavelengths for three CO₂ transitions [viz. P (20) and R (18) of the 00°1 - 10°0 transition and R (22) of the 00°1 - 02°0 transition] have been measured to one part in 10⁶. The refractive index of dry air has been measured at the above three wavelengths and found to be 1.0002688 ± 0.0000003.

Synchrotron radiation is being investigated as an alternative standard to black-body radiation for establishing standards of spectral emission, particularly in the ultraviolet. Initial measurements, in which the radiation has been compared with existing standards, have shown good agreement between the measured and predicted distributions over the range 350 to 600 nm. It is hoped to use the radiation to calibrate deuterium lamps, initially down to 200 nm, for use as working standards.

The boiling point of neon in terms of the NBS 1955 scale of temperature is found to be 27.0961 ± 0.00023 K. The experimental procedure is described, together with experiments to investigate random and systematic sources of error in the determination. It is concluded that this realisation provides a repeatability of ± 2.5 × 10^-4 K and an absolute error in a single observation of ± 3.3 × 10^-4 K. The result indicates that there is no serious shortcoming in the IPTS-68 assigned value of 27.102 K.