Among actual problems related to the measurement of macrostresses by X-ray methods we shall distinguish: metrological characteristic of these methods, measurement of elastic constants of polycrystalline materials (including textured materials), measurement of components of a stress tensor, measurement of gradient of a stress tensor in surface layer of polycrystalline materials. Let us consider the first one out of the above problems.
In all experiments two fundamental elements may be distinguished: subject of the experiment and the experimental method. Despite the importance of experimental method in every experiment, in general the analysis of its possibilities and their usefulness in real experimental conditions is underestimated. This comment refers also to X-ray stress method. Thus it seems reasonable to comment on at least such properties which play an important role in the selection or estimation of the method in the projection of the experiment or even in looking for the source of failure in inaccurately projected or realized experiments. The properties are characterized first of all by the following parameters: accuracy and precision of the method, its sensitivity, detectability of physical quantities, reproductibility and repeatability of results of measurements of determined quantities, efficiency and velocity of X-ray stress methods. In this paper the above mentioned parameters of X-ray methods in polycrystalline materials are described and analyzed. Different X-ray stress methods are considered. For those X-ray stress methods a basic cybernetic concept was used. In this way we obtain a cybernetic model of investigation method.
Experimental investigations show that precision of X-ray stress methods, characterized by a standard deviation values, are between 5 MPa and 18 MPa.
The upper boundary of an absolute error can be assumed to be triple the value of standard deviation. The values of this error for steels investigated are between 15 MPa and 53 MPa which indicates the possibilities of this methods.
Obtained expressions enable to estimate the sensitivity of chosen method in real experimental conditions. For example for Glocker-Hess-Schaaber method a maximal sensitivity is obtained for $\psi = 45^{o}$. In the previous experiments it was found that in this case measurement error is minimal. Both these facts account for the usefulness of this $\psi$ value in measurements.
Calculations for a carbon steel lead to following conclusions:
a) stress detectability depends really on the value of the standard deviation of diffraction angle: at constant stress its value increases as the deviation increases,
b) the values of sectioned-accuracy characterization depends on the stress and grows a little by its increase.
X-ray stress methods are easily reproducible. Repeatability of stress measurements at stable experimental conditions is good.
The analyses performed show that the X-ray stress method realized by conventional X-ray diffractometers is slow and has a low efficiency rate, however by using special diffractometers it becomes an effective and quick method.
This research was supported by Poznan University of Technology under Grant No. DPB2.