Results of the regional key comparison EUROMET.M.P-K1.a in the pressure range from 0.1 Pa to 1000 Pa

Within EUROMET a regional key comparison (EUROMET.M.P-K1.a) was performed in order to compare national vacuum standards in the pressure range from 0.1 Pa to 1000 Pa. The participants were BNM-LNE (France), CEM (Spain), OMH (Hungary), IMGC-CNR (Italy), NPL (United Kingdom), MIKES (Finland), PTB (Germany), NMi (The Netherlands), SP (Sweden) and UME (Turkey). IMGC-CNR acted as pilot laboratory. The measurements were carried out from November 1998 to April 2002. The chosen pressure values (from 0.1 Pa to 1000 Pa) cover the most commonly required range for calibration in low-pressure applications. The transfer standards were commercially available capacitance diaphragm gauges (CDGs): one of them was prepared for the comparison by BNM-LNE (Fr) and two were prepared by IMGC-CNR (It). Two sensors had 133 Pa full scale (one absolute and one relative used as absolute) and one 1333 Pa full scale (absolute). For the two 133 Pa full scale sensors seven pressure steps were generated between 0.1 Pa and 100 Pa; for the 1333 Pa full scale sensor nine pressure steps were generated generally between 0.1 Pa and 1000 Pa. The uncertainty of the generated pressure was reported by each participant in the tables of the results that consisted in the generated pressure value, the uncertainty of the generated pressure, the reading of the gauge and the temperature of the standard at each target pressure. The pilot laboratory has analysed the results, after application of the correction for thermal transpiration, in terms of gauge factors for each gauge, and the combined uncertainty was evaluated by considering, besides the component due to the standards, taking into account the components due to the transfer standards to guarantee a uniform uncertainty analysis for all the participants. At each target pressure a EUROMET reference pressure was calculated; finally the difference between the pressures generated by each laboratory from the reference values was calculated and compared with its expanded uncertainty. The results of most of the laboratories showed a good agreement with the reference values. Only a few values of two laboratories were significantly off the reference values. From the available data a linkage to the CCM.P-K4 key comparison results from 1 Pa to 1000 Pa was possible by means of the results of three laboratories in the (1–30) Pa range and two for the (100–1000) Pa range who took part in both the comparisons.