How to Achieve a 0.01 ìV/V Deviation on Your 10 Vdc Proficiency Test Without Using a Josephson Array

Abstract

How to Achieve a 0.01 ìV/V Deviation on Your 10 Vdc Proficiency Test Without Using a Josephson Array. This paper describes the steps the National Instruments Metrology Laboratory took to establish, in a short period of time, a sub ppm uncertainty 10 Vdc reference standard. One and a half years after the purchase of the 10 Vdc Zener references, the Proficiency Test (PT) reported an error of 0.01 ìV/V. The National Instruments Metrology Laboratory bought four 10 Vdc voltage reference standards in order to establish the 10 Vdc reference for the corporation and to enable the in-house artifact calibrations on calibrators. As a new laboratory, we could not benefit from historical data since our references were new. A primary goal of the project was to perform artifact calibrations one year after the purchase of our reference standards. Starting without historical data, the laboratory was able to achieve an error of 0.43 ìV/V on the first PT, just six months after the references were put into service. Why perform a PT after just six months using new voltage references? The drift of the reference standard is specified at 2.0 ìV/V for 12 months. The 10 V reference used for the artifact calibration must have an uncertainty of 1.5 ìV/V or less in order for the calibrator to meet its published specifications. A six month calibration interval for the 10 V reference allowed the uncertainty to be low enough to keep the calibrators within their specifications. This paper covers the program developed by National Instruments and the lessons learned that might help a new laboratory to make their progress quicker and cheaper. The most recent PT result is included, two and a half years after the purchase of the reference standards.