Progress Report on an International Intercomparison in the Pressure Range 10 kPa to 140 kPa

After the CCM Medium Pressure Working Group intercomparison indicated that the SMU primary mercury manometer differed by more than 20 ppm from other national standards, attention was paid to determining and eliminating the possible sources of these deviations. These efforts included replacing or redesigning some of the manometer components. The modified manometer design is presented in this paper. A piston gauge has been calibrated in the absolute mode using nitrogen as a pressure media, with the primary standard manometers at four national standards laboratories: Slovensky' Metrologicky' Ustav (SMU), Bratislava; Amt fir Standardisierung, Messwesen und Warenpriifung (ASMW), Berlin; Physikalisch-Technische Bundesanstalt (PTB), Braunschweig; and the National Institute of Standards and Technology (NIST), Gaithersburg; over a period of six years. Each of these manometers is of a different design. This has provided a basis from which to determine how the design changes affect the SMU manometer performance. The results indicate that the previous systematic deviations have been eliminated in the upper pressure ranges. The results for all four manometers agree to within their claimed uncertainties, however at the lowest pressures significant differences still exist and require further investigation.

Section: Intercomparison Resultsmentioning

confidence: 99%

“…Over a period of six years the manometer has been partially rebuilt to eliminate the systematic difference which was found during the CCM Medium Pressure Working Group intercomparison [ 2 ] . The original laser with a hot cathode was replaced with a cold cathode helium-neon laser with a separate interferometer unit.…”

Section: Introductionmentioning

confidence: 99%

The SMU Primary Mercury Manometer and its Comparison with Three Manometers of Different Design

Farár¹,

Skrovanek²,

Faltus³

et al. 1994

“…The transfer standard is a pressure balance specially adapted for that purpose. The results obtained to date, as the comparison is still going on, are presented by P. R. Stuart in another contribution to this section [8].…”

Section: History Of Intercomparisons Realizedmentioning

confidence: 94%

Pressure Intercomparisons at the Lowest Level of Uncertainty: Transfer Standards and Results

Legras

1994

The intercomparison is the best tool for the evaluation of the quality of a standard, and the validity of the estimation of its uncertainty. It is also the only way to detect possible systematic errors, experimental errors, missing items in the uncertainty budget or mistakes in calculation. Since 1975, many intercomparisons have been organized in the whole pressure range, all over the world. It would be a very large and difficult task to list them. Only the most significant are given as examples. The most recent are presented in other papers at this seminar. A review of the most commonly used transfer standards and their characteristics is given. The more significant data obtained from the intercomparisons (technical data and direct exchanges) are analysed.

“…Due to the high density of mercury, Hg 13 546 kg/m 3 , and the fact that most metals amalgamate with or dissolve in mercury, only very few materials can be used for sinkers intended to be employed for determining the density of mercury by hydrostatic weighing. Formerly, tungsten carbide was often used for this purpose [6, 9, 10], but as this material contains cobalt it is magnetic [6,9,10]. Also, cobalt dissolves slightly in mercury [11].…”

Section: Sinkersmentioning

confidence: 99%

“…Uncertainties (for a coverage factor ) of 0.1 Pa to 0.4 Pa near 101 kPa are claimed for modern mercury manometers [5], but the last comparison of pressures in the range 10 kPa to 120 kPa organized by the Consultative Committee for Mass and Related Quantities (CCM) revealed discrepancies of up to 3 Pa [5]. Although the densities of most of the mercury samples used were not measured, the discrepancies are probably not caused by density differences [6]. Nevertheless, in order to ensure that the claimed uncertainties are achieved, the mercury densities should be measured [7].…”

Section: Introductionmentioning

confidence: 99%

New apparatus for measuring the density of mercury

Bettin

Krumscheid

1999

A new apparatus is being set up for measuring the density of mercury at 20 C with a relative uncertainty of 0.5 ϫ 10 -6 by hydrostatic weighing of a sinker of known volume and mass. Two small sinkers of about 50 cm 3 were calibrated by hydrostatic comparison with silicon density standards. Thus, in the first stage, a relative uncertainty of 2 ϫ 10 -6 of the mercury density can be reached. The quantity of mercury needed for the measurements is 1 l. Also, density comparisons of different mercury samples can be carried out with a relative uncertainty of 0.5 ϫ 10 -6 . To reach a relative uncertainty of 0.5 ϫ 10 -6 in the density of mercury, a 500 cm 3 hollow cube is to be manufactured from tantalum. Its volume will be determined by means of an interferometer, with a relative uncertainty of 0.15 ϫ 10 -6 . Measurements will be performed for the gravitation experiment of Zürich University using 1 m 3 of mercury, and for mercury samples used in primary pressure standards. Design and results of first tests of the hydrostatic weighing apparatus are reported.