On uncertainty of the interpolating time interval meters

Abstract

A time interval meter (TIM) itself is perhaps the only measuring instrument in time and frequency metrology. It measures in a direct way the duration of time interval between two instants and expresses the measured value in adopted time units. In many applications especially in those which use continuously operating measuring systems with stable signals-it means signals with stable amplitude and frequency-for example in time laboratories, resolution of TIM is of a prime interest. The modern commercial TIM has reported resolution up to some tens picoseconds or even better. The most widespread method of resolution test is to apply across START/STOP inputs the same 1 ppS signals, usually originated from the same source as the reference clock signal. Sometimes users doesn't recognise that TIM is three ports instrument and interaction between input signals and clock one may influence not only on accuracy but resolution also. The aim of this paper is to show how to detect these crosstalks and to propose way to decrease it.