X-ray absorption techniques have become more common and more accepted in the study of moisture contents and transport in building materials like cement pastes, mortars, and wood. To be able to properly interpret the data gathered by the machine, it is necessary to understand how the machine s settings influence the resulting uncertainty of the measurements. Moreover, for larger specimens, it may be necessary to average data over several measurements in space or time to elucidate the mean features of the specimens. No detailed studies have heretofore been conducted to determine the uncertainty of X-ray absorption measurements or the appropriate averaging procedures. We evaluated an X-ray absorption system and found that the normalized root mean square errors (a measure of uncertainty) in X-ray counts decreased when the X-ray source intensity, the integration time, the horizontal resolution, the collimator size, and the number of scanning repetitions increased. The inherent uncertainty of the machine for a water specimen was about 2 % or less when the number of counts per specimen point exceeded about 17000. Averaging vertical count profiles for the water specimen reduces the vertical variability by half. For a paste specimen, when counts exceeded about 1500 per point, the uncertainty of vertical profile measurements is less than 2 % for any horizontal scanning resolution. For a mortar specimen, the horizontal spacing of vertical scans must be 2 mm or less with counts greater than about 4000 to have an uncertainty of 2 % or less. An integration time of 5 s produces uncertainties of 2 % or less when counts per second exceed 1000. Maximum normalized root mean square errors did not exceed 10 % in any of the tests conducted.
Citation: Journal of Research (NIST JRES) -
Volume: 109 No. 5
NIST Pub Series: Journal of Research (NIST JRES)
Pub Type: NIST Pubs
cement paste, moisture, mortar, non-destructive testing, Poisson process, x-ray absorption