X線応力測定法の精度向上に関する研究

抄録

The authors have pointed out in their previous papers that both the accuracy and reliability of X-ray residual stress value on the samples with narrow breadth of X-ray diffraction profile were sufficiently good. However, no results on the sample whose X-ray diffraction profile is very broad have yet been obtained. As the measurement of peak position of X-ray diffraction profile, although the half-value breadth method can be employed and good results were obtained in the case of sharp diffraction profile, this method cannot be applied to the case of broad diffraction profiles, because the back ground line of the profile cannot be clearly determined. Consequently, it is considered that the accuracy of the stress value obtained is not sufficiently good. Therefore, it is necessary to develop a new method for the measurement of the peak position of diffraction profiles.
In addition to this problem, the effect of Kα₂ component on the measuring stress ought to be considered. Furthermore, the effect of the Lorentz polarization factor and the absorption factor should be taken into account when the measuring sample has a large amount of residual stress with a broad diffraction profile.
In the present study, the authors attempted to separate the Kα₁ component from Kα doublet by using Fourier analysis so that this procedure may be applied to any shape of profile, and so they discussed the effects of the Lorentz polarization factor on the stress value.
For the determination of the peak position, first, two assumptions were made as follows:-1) The intensity distribution curve actually observed Φ(x) is given by the sum of the intensity distribution curves φ₁(x)and φ₂(x) diffracted from Kα₁ and Kα₂ radiations, respectively. 2) The following relation holds between φ₁(x) and φ₂(x),
φ₂(x)=kφ₁(x-Δx)
where k is a constant and Δx is the distance between the peak positions of the intensity curves diffracted from Kα₁ and Kα₂ radiations. Next, the intensity distribution curves Φ(x) and φ₁(x) were represented by the Fourier series respectively, and φ₁(x) was separated from Φ(x), and the Lorentz polarization factor was inducted into φ₁(x). Lastly, the peak position of intensity distribution curve obtained by the above manner was determined by the parabola fitting method. These calculations were carried out by the digital computer.
By using this method, the residual stresses on the some samples were measured, and on the basis of the experimental results, the following conclusions were derived.
(1) It is considered that the effect of Kα doublet can be ignored from the view point of the accuracy of stress value even if the value of residual stress is large.
(2) It is supposed on the other hand that the effect of the Lorentz polarization factor on the measured stress is remarkable especially for the specimen which has a broad profile. However, when the half-value breadth of the profile given by Kα doublet is less than 3°, it may be considered that this effect can be neglected.
(3) By separating the Kα doublet and then correcting the Lorentz polarization factor, the X-ray residual stress on the shot peened carbon steel specimen agrees fairly well with the mechanical one, compared with the X-ray residual stress obtained by the raw data.
(4) Although the effect of Kα doublet can be neglected, it is considered that this method has the characteristic of being suitable to the measurement of the peak position of diffraction profiles with whatever shapes including broad ones.