Table of contents

After a short review on the different attenuation measuring techniques, some special points of view in the application of the power method are described. By employing exactly measured standard attenuators in cascade, the modified measuring equipment can be used over a range of more than 100 dB. An error analysis is given considering all essential sources of systematical errors the limits of which are calculated. It is shown that the systematical error of the system described is less than 1 · 10^-4 dB for very small attenuation values. It is in the order of |δA| ≈ 1 · 10^-4 (1 + A) dB over the whole range, when A is the numerical value of the attenuation (in dB) being measured.

The absolute density of two heavy water samples rich in deuterium (with a grade higher than 99.9%) was determined with the hydrostatic method. The exact isotopic composition of this water (hydrogen and oxygen isotops) was very carefully studied. A theoretical estimate allowed to get the absolute density value of isotopically pure D₂¹⁶O. This value was found to be 1104.750 kg m^-3 at t₆₈ = 22.3° C and under the pressure of one atmosphere.

As an extension of the 4 πβ - γ coincidence technique, it is shown that for some radionuclides, the β count rate should be expressed as a function of at least two observed β efficiencies, and that the absolute disintegration rate can be obtained by a simultaneous extrapolation to 100% efficiency in the two or more dimensions. Measurements on ⁵⁷Co and ¹³⁴Cs are described as examples. For these nuclides, the separated efficiency functions are shown to be linear in theory as well as in practice. Thus a major source of systematic uncertainty in efficiency extrapolation, that due to the unknown form of the extrapolation function, is greatly reduced.

A method of data collection for the determination of absolute disintegration rates by the 4 πβ - γ coincidence technique is described, which gives nearly an order of magnitude smaller variance, for a given total counting time, than conventional methods. It is applicable when the β counter output is proportional to the β particle energy. The principle, which is demonstrated for ¹³⁴Cs, is to record the amplitudes of all the β pulses, followed by integration over variable limits. Expressions for dead time and resolving time corrections are derived.

The establishment of the NPL viscosity scale using glass capillary viscometers is described and the values obtained for NPL viscometers on subsequent recalibrations over a period of some years are given. The results of comparisons with independent viscosity scales in other countries are also tabulated.

From the results obtained on recalibration of the NPL viscometers and the comparisons made with other independent laboratories, it is concluded that the present scale is realized with an uncertainty of about 0.2% at low viscosities (5 cSt) rising to about 0.5% at high viscosities (5000 cSt).

A new measurement of the thermodynamic temperature interval between the silver and the gold freezing points with an infrared pyrometer has been carried out. The value of the freezing point of gold, referred to the value of the silver point on the IPTS-68, thus obtained, is 1337.41 K (σ = 0.059 K). The ratio of the spectral radiances at the two points is 2.4428 (σ = 1.2 · 10^-3) at the mean effective wavelength λ_e = 0.9980μm (σ= 0.3 nm). The resulting temperature interval is 0.17 K shorter than the IPTS-68 interval between the same points.

This paper compares measurements of the wavelength of an NPL constructed iodine stabilized laser, made in terms of the ⁸⁶Kr primary standard which were carried out at the BIPM, with an earlier measurement at NPL. The result of the BIPM measurements for the i component of ¹²⁷I₂ is λ (¹²⁷I₂, i) = [632991 399.7 ± 0.2] fm and differs by only 0.7 fm (1 × 10^-9) from the NPL result made on the same laser. The results provide an illustration of the reproducibility of the ⁸⁶Kr standard which is now considered to be of the order of 4 × 10^-9, and also illustrate the ability of the two laboratories with their different measurement techniques to perform high precision wavelength measurement. Some preliminary beat frequency measurements of the frequency reproducibility of BIPM and NPL lasers are also reported.