ASTM-B567:98(R2021) pdf free download – Standard Test Method for Measurement of Coating Thickness by the Beta Backscatter Method.
6.4.1 In the logarithmic range, the relative measuring error is nearly constant and has its smallest value.
6.4.2 In the linear range, the absolute measuring error, expressed in mass per unit area or thickness, is nearly constant, which means that as the coating thickness decreases, the relative measuring error increases. At or near = 0.35, the relative errors of the linear and logarithmic ranges are about the same. Thus, the relative error at this point may, for most practical purposes, be used to calculate the absolute error over the linear range.
6.4.3 In the hyperbolic range, the measuring error is always large because a small variation in the intensity of the beta backscatter will produce a large variation in the measured coating thickness.
6.4.4 For instruments that indicate only hackscatter count rate and not thickness directly, the count rate is normally converted to a thickness by means of an appropriate graph. Such graphs are generally valid only within a specific range of coating thicknesses so that extrapolation of a linear range calibration curve (straight line on rectangular coordinates) into the logarithmic thickness range will result in measurement errors. Similarly, extrapolation of a logarithmic range calibration into the linear thickness range will also produce significant errors. Many instruments that indicate coating thickness directly are limited to the combined linear and logarithmic coating thickness ranges but will be in error if measurements are attempted in the hyperbolic thickness range. The instrument manufacturer’s instructions must be followed relative to the limiting coating thicknesses beyond which the particular instrument being used may give substantial errors.
6.5 Resolving Time of the Detector—Because of the dead time of Geiger-Muller tubes (see 2.1.8), the number of pulses displayed by the readout instrument is always less than the actual number of backscattered beta particles. Normally, this does not diminish the measuring accuracy significantly unless the count rate is so high as to saturate the detector.
6.6 Source Geonzetrv— The greatest measurement precision is obtained with the source placed in a particular position with respect to the test specimen. This position depends on the collimation of the beam of beta particles from the source and the location, form, and size of the aperture. If possible, most of the beta particles emitted by the source should he hackscattered from the test specimen, and not from the aperture plate(n). The instructions furnished by the manufacturer of the instrument for mounting the source shall be followed exactly.
6.7 Curvature—This test method is sensitive to the curvature of the test specimen. However, the normalized backscatter curve remains nearly the same if the surface of the test specimen does not protrude into the aperture of the platen by more than about 50 pm. By the use of specially selected aperture platens or masks where the isotope is premounted in a fixed, optimum position, it is possible to obtain nearly identical readings on both flat and curved specimens. This permits the use of flat calibration standards for the measurement of curved specimens. The relationship between maximum aperture size and specimen surface curvature is peculiar, in most cases, to the individual instrument design. These details are therefore best obtained from the manufacturer’s data.