Conclusions
The orthogonal-regression-line factors have a nonlinear relationship to the recording scale, and therefore, the median plotted on the profilogram does not correspond to that of the profile. The error in determining the surface-roughness parameters is eliminated by using the formulas (7), (8), and (9). The median of a stationary normal process plotted by means of the least-squares method differs but little from the orthogonal regression line even if the amplitude of waviness is three times greater than that of surface roughness. Moreover, this median's factors have a linear relationship to the scale factors and require a smaller number of mathematical operations for their evaluation. Such a median plotted on the profilogram strictly corresponds to the profile's median.
Literature cited
Yu. V. Linnik and A. P. Khusu, Engineer's Handbook of the Academy of Sciences, USSR, 154-15D, No. 20 (1954).
GOST 2789-59, “Surface roughness.”
Yu. V. Linnik, Least-Squares Method and Foundations of the Observations-Processing Theory [in Russian], Fizmatgiz, Moscow (1962).
Operating Instructions No. 176 on the Processing of Surface-Roughness Profilograms, Standartgiz, Moscow (1960).
A. P. Kartasheva and N. I. Freidgeim, Izmeritel'. Tekh., No. 6 (1968).
P. E. D'yachenko, V. É. Vainshtein, and Z. P. Grozinskaya, Methods for Testing and Standardizing Wavy Surfaces [in Russian], Standartgiz, Moscow (1962).
I. V. Dunin-Barkovskii, Piezoelectric Profile Meters and the Measurement of Surface Roughness [in Russian], Mashgiz, Moscow (1961).
Additional information
Translated from Izmeritel'naya Tekhnika, No. 2, pp. 84–85, February, 1973.
Rights and permissions
About this article
Cite this article
Girshfel'd, L.Z. Application of the least-squares method in processing profilograms. Meas Tech 16, 294–297 (1973). https://doi.org/10.1007/BF00812584
Issue Date:
DOI: https://doi.org/10.1007/BF00812584