The automatically step-scanning diffractometer, which is provided with an X-ray integrating counting rate computer, a digital printer and counter step-scanner, is described in this paper. Most of the industrial materials for the X-ray stress measurement weaken the peak intensity. For instance, the heat treated materials broaden peak-profiles, and some of the machine parts, which have curved surfaces, limit their X-ray irradiation area to satisfy the X-ray focusing condition. In these cases, it is preferable to integrate the weak diffracted X-rays. On the other hand, it takes much more time to record one peak profile by the step-scanning method, than by the continuously scanning method. But when the peak position is determined by Koinstinen's method, it need not record the whole peak profile. Moreover, good accuracy can be gained.
A few results of the stress measurement of heat-treated materials, restricting the X-ray irradiation area to 2×2mm², are also presented.

View full abstract

It is generally thought that due to the coarse grains (left on the heat affected zone of a weldment), there is loss of accuracy in the measurement of the stress obtained from these regions by the X-ray diffraction technique. In the present work the specimens were oscillated for some milimeters in order to increase the number of grains concerned in the diffraction. Using the specimens cut from an electroslag-welded plate, the relation between the grain size and the oscillating stroke of the specimen has been investigated.

View full abstract

An X-ray film unit is described for measuring residual stress in large parts. The tube and camera are supported and rotated by an adjustable mechanism on a large table capable of supporting large steel parts. Cameras are used which have cylindrical cassettes. This enables a more convenient application of the technique of fitting a parabola to intensities across a diffraction peak to accurately determine the peak maximum. Intensities are corrected for angle-dependent factors. A theoretical stress factor is then used. Data is included and compared with the diffractometer.

View full abstract

In the X-ray stress measurement using a parallel beam diffractometer, the X-ray diffraction profile was approximately calculated by digital computer, assuming some setting errors, i.e. the offset of the specimens or incidental X-ray beams against the center of the scanning circle of X-ray detector. The incidence angle φ used were 0°, 15°, 30° and 45°. The effects of setting errors on the accuracy of stress determined by the sin²φ method were examined using peak angles of the calculated diffraction profile.
The relations between the peak angle and the offset of the specimens, obtained from the calculated diffraction profile, showed good correlation with those obtained from the experiment. The calculated diffraction peak angles, when the offset of the specimens or incidental X-ray beams is assumed showed cyclic changes in relation to the divergence angle α of the receiving soller slit.
It seems unavoidable, even under the most careful setting, that the error of the stress of Fe {211} determined by Cr-Kα amounts to ±2kg/mm², where α=0.35°.
Provided, on the contrary, that the error of the determined stress is allowable up to ±2kg/mm², the offset of the specimens or incidental beams against the center of the scanning circle is acceptable up to ±4mm or ±3mm respectively.

View full abstract

On the measurement of residual stress in the quenched steel and in the tempered steel, comparative studies have been made both of the X-ray method and the mechaincal method, differing in the measurement procedure and in the principle, and their results have respectively been investigated. After the plates or the cylinders were removed in succession from the outer layer, the resultant strain has been calculated from the surface residual stress measured by the X-ray method in the remaining part during the removal process. Then it is compared with the bulk strain obtained by the measurement of the residual stress made by the mechanical method.

View full abstract

The correlations between the change of stress Δσ_X measured by CrK_α radiation and the nominal stress Δσ_N were investigated on shot-peened spring steel specimens in various surface conditions.
Tests were made under uniform bending. Besides, the residual stress distributions in shot-peened decarburized or non-decarburized specimens were examined.
In case of shot-peened decarburized specimens, the ratio Δσ_X/Δσ_N was 0.59, and the residual compressive stress measured by X-rays showed steep ascent in the surface adjacent layer up to 30μ depth. The peak of the residual stress distribution appeared at about 100μ depth from the surface, and gradually decreased thereover. When the surface layers of shot-peened decarburized specimens were electrolytically polished to 100μ, the ratio Δσ_X/Δσ_N attained to 0.86∼0.88. When the residual stress of shot-peened decarburized specimens were relieved by annealing, the ratio Δσ_X/Δσ_N became 0.73∼0.82.
In case of shot-peened non-decarburized specimens, the ratio Δσ_X/Δσ_N was 0.84∼0.93, and residual compressive stress measured by X-rays remained almost constant in the surface adjacent layer.
It has been confirmed that the stress of shot-peened specimen measured by X-rays is much affected by steep residual stress distribution in the surface adjacent layer, but scarcely affected by the surface roughness.

View full abstract

The X-ray back diffraction method was used for the nondestructive direct detection of fatigue damage in the machine parts of larger size. The size effect on fatigue of annealed carbon steel was also studied.
The results obtained in this experiment are as follows.
(1) Whether the applied cyclic stress is larger or smaller than the fatigue limit of the specimen can be estimated by the back diffraction pattern in the case of unnotched test specimens, irrespective of the specimen sizes. On the other hand, the fatigue limit estimated from the X-ray pattern in the case of the specimen with a shoulder fillet is a little lower than the true fatigue limit. It is considered to be due to the stress gradient in the fillet.
(2) The changes in residual stress and half-value breadth of diffraction line width during the fatigue process in the case of unnotched specimens depend on the specimen size even under the same applied stress. This result shows also that the size effect on fatigue strength must vary according to the strength near the specimen surface, which will be emphasized by stress gradient.
It can also be concluded from the results obtained that the X-ray diffraction method is available for the detection of fatigue damage in the specimens of different size.

View full abstract

The change of the surface residual stress of the hardened ball bearing steel which was frictionally worn out under boundary lubrication was investigated. A slider with various kinds of materials was held for two minutes against the periphery of the ball bearing steel disk which was rotating at a velocity of about 1600m/min. The materials used for the slider were classified in four groups, such as (1) steels with and without surface coating, (2) pure metals, (3) nonferrous alloys, (4) plastics.
The residual stress was measured by the X-ray method, using a Geiger counter diffractometer with CrKα radiation.
The residual stress left to the rubbed surface of the disk was various according to the kinds of the materials of the slider. In general, when the slider is of good scoring characteristics (for example, plastics and some nonferrous alloys), the residual stress was unchanged or turned to much more compressive than the original value of -25∼-30kg/mm², while, for the sliders of poor scoring characteristics, the surface residual stresses of the disks changed to nearly zero or tensile.
In addition to the residual stress change, the half height breadth of diffraction line was also observed as decreasing. Since the half-height breadth affect the fatigue strength of steel components, these values would be used as criteria for the scoring characteristics of the materials sliding on the hardened steel.

View full abstract

The microstructural change around a fatigue crack in 15% cold-rolled low-carbon steel specimens was investigated by using the back-reflection X-ray microbeam diffraction technique. The results obtained are summarized as follows:
(1) Dislocations introduced by predeformation rearrange themselves in the vicinity of a fatigue crack. The excess dislocation density in a subgrain, the subgrain size and the micro lattice strain in a grain are decreased by this rearrangement of dislocations while the excess dislocation density at subboundaries increases. In this way the preformed substructure developes near a fatigue crack. Micro cracks are expected to exist at subboundaries of fatigue-induced substructure and a main crack will propagate along subboundaries, joining these micro cracks.
(2) The crack propagation rate has a great deal to do with the degree of development in substructure at crack tips, that is, the amount of decrement in micro lattice strain and subgrain size become larger as the propagation rate increases. It is considered that by using this relation, the estimation of the propagation rate of a crack whose rate is unknown will be possible, if the information about substructure at the crack tip is given in the observation by the X-ray microbeam.

View full abstract

The structural change of Al crystals due to the cyclic stress in bending was examined by using the back reflection divergent X-ray technique.
Al crystals, 99.999% in purity, were made by the strain annealing method. Of crystals thus obtained those of the following three kinds of specimens were used: a bicrystal and two single crystals having the same orientation as those of each part of the bicrystal.
The back reflection divergent X-ray photographs of fatigued Al crystals showed the occurrence of polygonization in the early stage of fatigue, but this structure was changed into complicated one with the progress of fatigue process. The fatigue life in the bicrystal was found to be shorter than that in the single crystals, and to form fatigue crack along the crystal boundary.
The same experiments were carried out with polycrystalline samples. It was found clear that the degree of deformation varied with the grains, some grains being deformed scarcely even in the later stage of fatigue process. The X-ray photographs obtained from the various stages of fatigue process showed a rapid increase of strain, especially, near the crystal boundary.

View full abstract

When the mean stress is applied to the notched low-carbon steel specimens subjected to alternating push-pull load, the effect on the crack propagating rate is discussed from the view point of the microstructural change at the crack tip observed by the X-ray microbeam diffraction technique.

View full abstract

It is generally known that the tufftrided steel has a high fatigue resistance, and that the improvement of its fatigue property must depend on the residual stress on the surface and the formation of nitriding layer there.
In this paper, as a part of studies in series on the fatigue fracture of tufftrided steel, the changes in the microstructures and residual stresses due to the reversed stress were observed, and the results obtained are discussed.
Based on the experimental results, the conclusions are summarized as follows:
(1) The number of slip lines due to cyclic stress is less than in the non-tufftrided steel specimen.
(2) The plastic deformation due to the fracture is limited to the region nearest the fractured part on the surface.
(3) No change has been observed in the residual stress on the surface of the layer during the fatigue process.
(4) When the nitriding materials are precipitated on the diffusion layers by tempering, their endurance limit decreases as compared with the case where there is no precipitation on the tufftrided steel, and there is scattering of intensity of the materials variable with time.

View full abstract

To investigate the effect of surface work hardened layer on the completely reversed bending fatigue strength, a 0.17% carbon steel (JIS SS41) was tested. The specimens were machined and their surfaces were ground, and vacuum annealed at 650°C for 45 minutes. One group of specimens was electrolytically polished to remove the surface work hardened layer, and the other group was shot peened and the surface roughness was smoothed, to some extent, by emery paper and electrolytic polishing. The fatigue tests were carried out with the Schenck type machine and the change of residual stress was measured during the fatigue process by the X-ray method.
The S-N diagram shows that the shot peened specimens have higher fatigue limit, but at the high stress amplitude the fatigue lives are shorter than the specimens without the work hardened layer. The surface compressive residual stresses of the shot peened specimens are decreased to 2/3 of the initial value by application of only one cycle of stress amplitude. After that, however, the decrease of residual stress is not so drastic but linearly proportional to the logarithm of number of stress cycles. But when the stress amplitude is below a critical value the residual stress is changed only by the first stress cycle, and remain unchanged thereafter.

View full abstract

Carburizing is a favored process to improve the fatigue strength and wear resistance of automobile gears and many other machine parts. Despite many years of studies and reports on the properties of carburized hardened steels, there still remain many problems to be solved in this respect where production is concerned.
The values of surface hardness, core hardness and case depth are known as main factors which influence the strength of carburized hardened steels. Among these, case depth and core hardness are variable extensively according to the differences in chemical composition, hardenability, and methods of carburizing and quenching, even if the parts are of the same quality. Moreover, those values vary widely depending on the hardenability of steels, even when the same type of steel is treated at the same temperature.
There have been many reports in which comparisons are made among the types of steels and the methods of carburizing and hardening but very few reports are obtainable concerning the investigations that dealt with these two factors at a time, especially on one type of steel.
Therefore, we investigated the influences of case depth, core hardness and residual stress on the fatigue strength of carburized hardened steels. The materials tested are chrome-molybdenum case hardening steels (SCM 21 & 22) of various hardenabilities.
The following points were made clear as the results of these tests. The fatigue strength of carburized hardened steel is influenced by the case depth and core hardness, both of which have optimum values to cause the maximum endurance. When the core hardness is low and the case depth is thin, crack initiates from case-core boundary (fish-eye). The depth of this fish-eye from the surface is almost equal to the total case depth. The decrement of residual stress during the fatigue test is rather small, and comparatively large residual stress is observed even after the test.
The surface residual stress of carburized hardened specimens first increases, then tends to decrease as the case depth is increased, and this can be considered to be one of the main causes for the behavior of maximum endurance of carburized hardened steels.
From those results, it seems necessary that further studies are pursued to ascertain the effect of residual stress, in order to attain high fatigue strength even for production parts.

View full abstract

The change in the residual stress induced on the surface of unnotched mild steel specimens by turning was investigated under rotating bending fatigue tests. The residual stresses were measured periodically on the surface by the X-ray film method.
The fraction of decrease of residual stress to the original residual stress vs. the logarithm of the number of cycles were found nearly linear, and the slope of these lines depended upon the applied stress-amplitude. At stress amplitude near the yield strength of this material, i.e. σ=40kg/mm², there was remarkable relaxation of residual stress. At this stress amplitude, the initial residual surface stresses of about 35∼40kg/mm² induced by turning was reduced almost to zero after the number of cycles N=10³∼2×10³ (cycle ratio was about 0.1∼0.2).
The S-N curves of the specimens with and without work hardened layer, annealed before and after turning, crossed each other at high stress level. At least, as one of the reasons for this fact it can be considered that the beneficial compressive residual stresses which work efficiently at low stress level faded away in the early stage of the life at high stress level.
Usually we use rotating bending fatigue specimens machined by turning. Then, in multiple stress amplitude fatigue tests, especially with the initial stress level higher than the second stress level, the decrease of residual stress must be taken into consideration.

View full abstract

The X-ray observation with diffraction technique was performed of microstructural changes in 0.16% carbon steel during constant cyclic straining of several amplitudes with mean strain at zero.
During low-cycle fatigue, the integral breadth ratio of X-ray diffraction lines changed in a way different from that found in the case of high-cycle fatigue of an annealed carbon steel. That is, the larger the strain range was, the larger integral breadth it caused. And the series of cyclic straining showed tendency to decrease the integral breadth somewhere after the half-way mark to failure. The mode of change in the integral breadth was investigated by conducting a profile analysis of the diffraction lines.
The particle size and microstrain measured by the analysis were used as the parameters indicating the microstructural changes that occurred during the low-cycle fatigue. And some trial was made to correlate the microstructural changes with a damage fraction. Further studies along these lines are expected to yield significant results in the field of low-cycle fatigue studies.

View full abstract

A compression X-ray camera is described in the present paper, devised for obtaining an X-ray photograph by transmission technique for the crystalline high-polymeric materials under uniaxial compression up to 2000kg/cm². The compression is applied by a press screwed by a manual worm-gearing which are set on the camera. The compressive force can be measured by a special gauge placed in the camera in series under the sample holder, and it can be kept constant, when desired for the stress constant experiments, by driving the press during the exposure to compensate the ralaxed stress of the sample.
A technique for an accurate measurement of the small radial shift of the diffraction peak induced by the compressive stress is also described where a flat film cassette equipped with a special metal shield which permits exposure of 2°C section of the film is used. The shift was measured by a direct comparison of the shifted peak under a microscope with the peak obtained under no applied stress and registered, by the use of the metal shield, on the same film in the immediate neighbor of the shifted peak.

View full abstract

To obtain the accurate lattice parameter of material by using Debye-Scherrer camera, it is necessary to consider the various effective factors which can cause errors (systematic error) in the measurement of Bragg angle θ, and to find out how these errors in θ vary with the angle itself. In practice, the effect of thickness of the specimen on the measurements of θ is very important in finding the accurate lattice parameter. Nishiyama proposed a method for eliminating this effect on the assumption that an X-ray source is large enough to the radius of the specimen. His assumption, however, becomes incorrect when the radius of the specimen is not negligible compared with that of the X-ray source.
The purpose of this study is to consider the error in the value of θ due to the size of the X-ray source, and to compare the results obtained by the proposed method with those obtained otherwise. It can be concluded from these experiments that the size of the X-ray source is an important factor to obtain the accurate lattice parameter of the material under study.

View full abstract

In a series of studies on the deformation mechanisms of polycrystalline metals, the experimental and analytical investigations on X-ray elastic constants are presented in the present paper. The diffracting plane dependence of X-ray elastic constants as well as the changes in these constants by plastic deformation are observed experimentally, and the interpretation of the results obtained is given on the basis of crystal orientation, anisotropic elastic constants and deformation texture.

View full abstract

In this experiment, residual stresses and half-value breadths were measured by the X-ray method, on the specimens of carbon steels with circular notches, by application of static or impact loads, followed by application of combined static-impact loads. For impact loading a simple apparatus of drop hammer type was used, and the applied compressive strains at the notch root (0.2 to 0.9%) were measured with a synchroscope. The X-ray measurements were carried out by the film method using CrK_α radiations at the notch root, and on the Lüders band at three locations.
The experimental results obtained are summarized as follows. In the case of separate loading, both residual tensile stresses and half-value breadths increased with the applied strains, and the increase in the latter was remarkably larger in the case of impact load. In the case of combined loading of the same kind, the increases in half-value breadth were smaller and, on the contrary, the residual lattice strains were larger as compared with the case of individual loading. On the other hand, in the case of combined loading of different kind, the change in the sequence of loading gave the inverse tendency. On the Lüders band, the residual compressive stresses as well as the increases in half-value breadth showed larger values at points located nearer to the end point. The etching experiment revealed that the residual stresses are restricted within the utmost thin layers from the surface, and that the region of lattice distortions is assumed to penetrate the deeper layers under impact load.

View full abstract

In this experiment, as sequel to the previous report, the residual stresses and half-value breadths were measured by the X-ray method on the specimens of carbon steels with a circular notch on one side, after the application of static or impact loads. The effects of annealing after loadings and grain sizes of the specimens were also investigated. For impact loading a simple apparatus of drop hammer type was used, and the applied bending strains at the notch root (0.2 to 1.0%) were measured with a synchroscope. The X-ray measurements were carried out by the film method using CrK_α radiations at the notch root. The grain sizes obtained by various annealing treatments were ASTM No. 9, 6, 4 and 3.
The experimental results obtained are summarized as follows. In the case of static or impact loading, both the residual compressive stresses and half-value breadths increased with applied strains, and the increase in the latter was conspicuous in the case of impact load. The comparison with the previous compression tests gave the suggestion that the lattice distortions such as bending are larger in impact bending than in impact compression. The experimental results of annealing showed that both decreases in the residual stress and the half-value breadth are faster for static than for impact load. The residual stresses almost disappeared after 500°C annealing, however, half-value breadths failed to recover their original values, especially for the case of impact loading. As for the effect of grain sizes, the residual stresses were smaller and, on the contrary, the increases in half-value breadth were larger for smaller grain sizes.

View full abstract

The creep behavior of materials under intermittent load is one of the most important and attractive problems in the field of physics and mechanics of engineering materials.
In order to investigate the creep mechanism of metallic materials under intermittent load at elevated temperature, it was intended to observe the structural change in 0.1%C steel during creep under static load, and the same change in the same material during creep under intermittent load, employing the X-ray diffraction technique.
As the results of this investigation, the following conclusions have been obtained,
(1) Deformation of transient creep which occurs during each reloading half-cycle is caused by the same mechanism, but this mechanism is found to be different from that of the transient creep in the case of the creep under static load.
(2) Half-value breadth of X-ray diffraction profile is found to be effective for characterizing the mechanical behavior of creep.
(3) The difference found in the half-value breadth change between the crept specimens under the static load and the same specimens under the intermittent load is correlated to the difference in the strain rate between these specimens.

View full abstract