embodiments, like glass-hemisphere and sensitivity standards [2-6].
There is only one type of form embodiment widely available which is applicable for the sensitivity calibration, the so-called flick standard which is a cylinder with a flat face. Nearly all roundness measuring instruments are calibrated by using a flick standard.
Previously, calibration of the roundness measuring machine at NIMT has been performed routinely using two flick standards (magnification setting standards), 19.76 m and 273 m, and a glass hemisphere. Glass hemisphere can be self-calibrated by using a multi-step method. However, flick standards have been calibrated by the national measurement institute of Australia (NMIA) because NIMT did not have calibration facility for flick standard with low uncertainty. This result in roundness measurement in Thailand is traceable to SI unit through NMIA. With mentioned calibration routine, roundness measurement at NIMT can be achieved with uncertainty of Q[11.1, 0.01R] nm at k = 2.
Many NMIs have developed calibration systems and new standard to evaluate sensitivity of the roundness measuring instrument instead of using flick standard [7-11]. Most systems are based on principle of laser interferometer due to its high resolution, high accuracy, high stability and direct traceability to the SI unit.
This paper deals with new calibration system for the sensitivity calibration. Aims of this work are to reduce calibration cost of flick standards by oversea NMIs, to increase accuracy of the roundness measurement at NIMT and to develop a calibration system that can be modified to various types of roundness measuring instrument.