Table of contents

An automatic tuning circuit for correcting the resonant frequency of the cavities used in a pair of hydrogen masers is described, with particular emphasis on measurements of the relative frequency stability obtained using the system. The measurements show that improved long-term stability of the two NRC masers can be obtained by the use of the auto-tuner. The stability of the measured frequency difference between the masers is now typically 1.5 parts in 10¹⁴ over periods up to one week. Correlation has been found between changes in the frequency corrections supplied by the auto-tuner and changes in atmospheric pressure, indicating that the cavity frequency of each maser is sensitive to pressure changes. The magnitude of this effect and modifications made to the masers to reduce it, are discussed.

The scale presented in this document was proposed by the International Committee of Weights and Measures in 1974 and adopted by the 15th General Conference of Weights and Measures in 1975. It constitutes only an amendment of the International Practical Temperature Scale of 1968 (IPTS-68), not a replacement. Any measured temperature, T68, remains unchanged by this amendment of the IPTS-68.

The freezing point of platinum has been measured on the International Practical Temperature Scale of 1968 to be 1769.5°C ± 0.6°C. The instrumentation and methods of experimentation are described. All significant uncertainty components are stated. Possible systematic errors are discussed and their magnitudes evaluated.

The low frequency (30 Hz) paramagnetic susceptibilities of powdered samples of gadolinium sulphate octahydrate and manganous ammonium sulphate hexahydrate have been measured from 1 to 83 K in order to test the thermodynamic consistency of the International Practical Temperature Scale of 1968 (IPTS-68). The data can be characterized by a physically based expression with essentially two constants as opposed to the twenty-one terms of the IPTS-68 Reference Function for the range 13.81 K to 273.15 K. Variations in the IPTS-68 of up to ± 3 mK about the smooth magnetic scale are found and the 20.28 K (e-H₂ b.p.) fixed point is shown to be inconsistent with the other fixed points by about +3 mK. The data also provide a means of thermodynamically extrapolating the IPTS-68 to lower temperatures where comparisons are made with other recent data. The reproducibility of the data is ± 0.5 mK below 50 K and ± 1 mK up to 83 K, while the total estimated measurement uncertainties are from one to three times greater.

A commercial laser with an intracavity etalon was fitted with an internal ¹²⁹I₂ absorption cell. A rather isolated ¹²⁹I₂ hyperfine component near the ArII line center is used as a reference for laser wavelength stabilization having been achieved by the third harmonic locking technique. The half-width (HWHM) of the hyperfine component is approximately 5 MHz. Therefore, the laser wavelength stability should be Δλ/λ = ± 10^-10. The stabilized laser wavelength was measured in comparison with the primary standard of length, yielding the value λ_vac = (514.67465 ± 0.00001) nm.