Investigation of the uncertainty contributions of the alignment of PTB's double-ended interferometer by virtual experiments

Abstract

High-accuracy length measurements of prismatic bodies (e.g. gauge blocks) are usually performed by means of single-ended interferometers. To perform these measurements, the gauge block must be wrung to a reference plate. The quality of contact affects the measured length and also the wringing process wears down or damages the measuring faces. Furthermore, it limits the use of such interferometers to bodies that are suitable for wringing. PTB's double-ended interferometer allows high-accuracy length measurements that are traceable to the International System of Units to be performed without a reference plate. However, because the setup of this interferometer is complex and additional optical components are required the alignment process is challenging. Compliance with the defined gauge-block length in ISO 3650 is also challenging, especially for non-perfect shaped gauge blocks. In this work, we develop a precise alignment method for the double-ended interferometer and systematically study the contributions of misalignments to the uncertainty of the measured length. In order to explore the accuracy of the developed procedure and to estimate the size of the uncertainty caused by deviations from perfect gauge block shapes, virtual experiments are carried out using the PTB library SimOptDevice. The virtual experiment is validated by a comparison to experimental data. In addition, theoretical relations are confirmed. Finally, Monte Carlo runs of the virtual experiment are performed to quantitatively explore the size of different sources of uncertainty on the developed alignment method. The results suggest that the developed alignment method is highly accurate and is expected to yield an uncertainty contribution to the final length measurement in the sub-nanometer range.