An Investigation of the Minimum Expected Uncertainty in the Linear Traverse Technique

The paper presents equations which permit the direct calculation of the uncertainty in estimating air void system parameters from the results of a single linear traverse. The uncertainties calculated and verified by computer simulation are free of operator or equimpent error which are normally present in a linear traverse analysis. The reported uncertainties are therefore inherent to the linear traverse technique. The study shows (analytically and experimentally) that the expected uncertainty depends strongly on the number of chords in the traverse. Additional uncertainty occurs for an increse in the 'spread' of the bubble radii distribution, corresponding to a 'spreading' in the chord distribution.