Supplementary comparison: Secondary calibration of accelerometers in medium frequencies
The supplementary comparison EURAMET.AUV.V-S1 has been carried out within the framework of the CIPM MRA. The specific task of this comparison is the measurement of the magnitude and phase of the sensitivity of single-ended accelerometers in the medium frequency domain (10 to 10 000 Hz). The sensitivity is calculated as the ratio of the amplitude of the output of the accelerometer to the amplitude of the acceleration at its reference surface with secondary means in accordance with ISO 16063-21 "Methods for the calibration of vibration and shock transducers - Part 21: Vibration calibration by comparison to a reference transducer". The participating NMIs are BEV (Austria), BIM (Bulgaria), CMI (Czech Republic), IPQ (Portugal), KEBS (Kenya), METAS (Switzerland) and MIKES (Finland). BEV (Austria) was the pilot laboratory of the comparison. The measurements took place between October 2019 and February 2021. Three single-ended accelerometers were circulated. The final report includes the measurement results from the participants, information about their calibration methods, and the analysis leading to the assignation of equivalence degrees.
Final Report 2022-03-11
The aim of the supplementary comparison EURAMET.AUV.V-S1 was to establish reliable equivalence between national metrology institutes for the calibration of accelerometers with secondary means according to ISO 16063-21 standard. Three single ended accelerometers were circulated among the participants. Due to its poor stability, one accelerometer had to be excluded from the analysis of the results. The results of primary calibrations according to ISO 16063-11 standard at METAS were used as reference values for the analysis of the results of the secondary calibrations.
There are two typical issues of the method of secondary calibration of accelerometers:
At high frequencies, relative movements can occur and disturb the calibration results. This effect is called payload effect or mass loading effect. The relative movement results in an apparent change of the sensitivity of the reference transducer. This effect increases with increasing mass load and frequency. It especially affects external back-to-back reference transducers. Typically, for frequencies above 5 kHz, it can result in significant deviations of the calibration results.
At low frequencies (< 20 Hz), increased harmonic distortions and signal noise can occur and disturb the calibration results. The signal-to-noise ratios of the transducer to be calibrated, and the reference transducer have to be observed and considered. In addition, large displacements may induce cable-noise via the triboelectric effect of the bending cable.
One aspect of the comparison task was the complementation of the measurement chain with a local charge amplifier calibrated by the participant. Due to this requirement, the consistency of the results of most participants in supplementary comparison demonstrates the associated capability of calibration of magnitude and/or phase of the complex transfer function of charge amplifiers. In addition, this capability could be expected to be valid for other kinds of measuring amplifiers as well, including voltage amplifiers.
Overall, the supplementary comparison EURAMET.AUV.V-S1 can be considered successful for most of the participants. The calibration of accelerometers with secondary means according to ISO 16063-21 standard is still offered by several metrology institutes. Its technical challenges may be less demanding compared to primary calibration, but the results of EURAMET.AUV.V-S1 show, that the method of secondary calibration is worthwhile to be addressed by a metrology-level comparison. After all, secondary calibration is the most common calibration method in accredited laboratories worldwide.
Progress Report 2020-08-11
October 2019: Initial measurements of the transfer standards completed at BEV
November 2019: Standards sent to IPQ, Portugal and VTT, Finland
February 2020: Remeasurement of the transfer standards at BEV
February 2020: Standards sent to CMI, Czech Republic and BIM, Bulgaria
May 2020 – June 2020: No measurements due to Corona lockdown
July 2020: Remeasurement of the transfer standards at BEV
August 2020: Standards sent to METAS, Switzerland
So far, at least two of the three transfer standards show sufficient stability for the comparison.