Long-term stability of metal-envelope enclosed Bayard–Alpert ionization gauges

Abstract

Ionization vacuum gauges are used as secondary standards by calibration laboratories and as transfer standards in intercomparisons among metrology laboratories. A quantitative measurement of gauge stability with respect to the gauge calibration factor is critical for these applications. We report the long-term calibration stability of hot-filament metal-envelope enclosed ionization gauges based upon the analysis of repeat calibrations of nine gauges over a 15 year period. All of the gauges included in the study were of the same type: Bayard–Alpert type ionization gauges of an all-metal construction with an integral metal-envelope surrounding the hot-filament, grid, and collector. All were calibrated repeatedly at the National Institute of Standards and Technology (NIST) using the NIST high-vacuum standard but are owned by organizations external to NIST. The gauges were removed from the high-vacuum standard after calibration, shipped back to the gauge-owner, and were returned to NIST at a later date (more than 1 year) for recalibration. Gauge stability was determined using a pooled standard deviation (weighted root-mean-square average of individual gauge standard deviations) based on all calibration factors measured at NIST and was used to define the relative uncertainty component associated with long-term stability uLTS. We determined uLTS = 1.9% (k = 1) for gauges operated with 4 mA of emission current, and uLTS = 2.8% (k = 1) for gauges operated with 0.1 mA emission current.