THE INFLUENCE OF GRAIN SIZE ON THE IMPACT-FATIGUE RESISTANCE OF STEEL 45

G. V. Toporov, G. Ya. Smokotin

The operating conditions of machines used for mining work impose especially high requirements on the durability of parts, owing to the difficulty of organizing repairs at the place where the machine is operating. The problem of increasing the durability of parts of mining machines, despite the efforts of factories and scientific teams, remains as yet unresolved.

It is important to note that a large number of machine parts (drilling and jack hammers, coal combines, etc.) during operation are subjected to repeatedly recurring impacts. In investigations we carried out on the structure of broken parts operating under conditions of impact cyclic loading, it was established that the overwhelming majority of them had a coarse-grained structure. The coarse-grained structure of broken parts (links of scraper-conveyor chains) was also noted by M. M. Tenenbaum [1, 2].

In the practice of mining machine building, in parts subjected to normalizing or annealing, or else not undergoing heat treatment at all, the grain size is, as a rule, not specified. Moreover, heat-treatment regimes are sometimes prescribed with the aim of better machinability of the parts, without taking into account the possible reduction in mechanical properties.

It is known that the introduction of the operation of hardening and tempering (quenching followed by high tempering) encounters especially strong objections among production workers, since in this case a deterioration in machinability by cutting is observed. Indeed, steel with a sorbitic structure, obtained after hardening and tempering, possessing greater hardness and poorer machinability, at the same time is distinguished by higher mechanical properties. In this connection there arises the question of the advisability of applying heat treatment—hardening and tempering—to parts operating under an impact-cyclic character of loading. This question can be resolved only after clarifying the advantages in resistance to impact-fatigue fracture of steel after hardening and tempering as compared with normalizing.

In the extensive literature devoted to the influence of grain size on mechanical properties and fatigue failures

steel, we have not encountered works on questions of the relationship between grain size and resistance to impact-fatigue fracture of this steel.

In the present work, experimental data are presented that were obtained in studying: a) the dependence of the resistance to impact-fatigue fracture of steel 45 on grain size and b) the influence of various types of heat treatment (refining, normalizing) on the durability of this same steel under repeated impact loading.

On Testing Metals by Repeated Impacts

It is known that steel with a coarse-grained structure has an increased tendency toward brittle fractures. The difference in properties between the coarse- and fine-grained structure of one and the same steel is revealed in dynamic tests (single impact) for impact toughness. At the same time, as has been demonstrated by numerous theoretical and experimental investigations of Academician N. N. Davidenkov [3, 4, 5, 6] and his school, almost all static tests give results that do not distinguish fine-grained steel from coarse-grained steel. This also applies to tests of the static fatigue of metals [7].

The strength of parts subjected to repeated (multiple) impact loads can be assessed either by the limit of impact endurance or by the number of impacts to fracture at a specified constant impact energy.

As N. N. Davidenkov and E. I. Belyaeva [8] point out, “metallurgists began to concern themselves with impact fatigue as early as 1908, when Stanton built for this purpose a primitive little machine,” on which a rotating cylindrical specimen with a notch was subjected to bending impacts. As a result of the analysis of the experimental data of Stanton and Berstow, McAdam, and others, the view arose at that time that “tests by repeated impacts simultaneously contain elements of two different kinds of tests: impact tests of notched specimens and tests for ordinary fatigue” [4], and that tests in repeated-variable impact bending give nothing new in comparison with these two kinds of tests. Moreover, the limit of impact fatigue was expressed not through the stress arising in the specimen during impact, but through the energy of a single impact, which ruled out the possibility of comparison with corresponding data for static fatigue. All this led to the discontinuation of impact-fatigue tests.

However, the use of high speeds and pressures in modern technology, including coal and mining technology, has noticeably broadened the range of machine parts that experience dynamic loads in the form of multiple impacts. Therefore, the views that had previously taken shape regarding tests by repeated impacts

were subjected to criticism and experimental verification. Despite their complexity and long duration, these tests began to be revived.

A prominent place in the development of the methodology, in the experimental-theoretical substantiation of testing metals by repeated impacts (bending, tension, compression), and also in the study of the process of propagation of an impact-fatigue crack belongs to Prof. A. I. Lampsi.

Speaking against one-sidedness in assessing the effect of repeated impacts on a part, Prof. A. I. Lampsi [9] writes: “...a very large number of machine parts and engineering structures, under alternating stresses, are subjected to repeated impacts of large impulse and high frequency, while, in setting up tests and accounting for the operating conditions of alternating stresses on a material or on a structural part, the indicated repeated impacts are not taken into account,” although these impacts significantly change the nature of the action of alternating loads.

With regard to the analysis and methods of eliminating the causes of failures of coal-machine parts, it may be said that there is a certain tendency to use the achievements of surface-hardening methods [1, 2]. However, these methods are expedient and effective when the broad possibilities of heat treatment for increasing the resistance of a given steel to repeated impacts have been completely exhausted. It seems probable that underestimating the possibilities of heat treatment may lead to making machines more expensive without substantially increasing their service life. Therefore, the introduction of testing steels by repeated impacts is dictated by the practical needs of coal-machine building.

There exists an opinion that if a high value of impact toughness has been obtained for a given steel, then a part made of this steel will resist repeated impacts well. In other words, universality is attributed to the mechanical characteristic of steel under a single impact—impact toughness. Apparently, on this basis there exists a tendency to use steel in the normalized condition, since it has greater impact toughness in comparison with the impact toughness of the same steel after quenching and high tempering (improvement).

Prof. G. I. Pogodin-Alekseev [10] notes that, in addition to a high capacity to absorb kinetic energy, the material of parts operating under impact is also required to have high surface hardness, which enables the part to withstand local deformations at the point of impact. In other words, the resistance of steel parts under impact cannot be determined solely by the value of impact toughness. As toughness increases, the value of hardness decreases, i.e., diminishes—

… the resistance of the part to plastic deformation and, conversely, an increase in the surface hardness of the part (especially a carburized one), may lead to exhaustion of ductility and impact toughness, and the part will be brittlely fractured.

Experimental data on the study of impact fatigue are few. Prof. N. N. Davidenkov and E. I. Belyaeva [8], using the advances of strain-gauge measurement, succeeded in measuring, in terms of stresses, the impact-fatigue limit for certain steels under various heat treatments and in comparing it with the static fatigue limit of the same steels.

N. F. Lashko [11] investigated the aluminum casting alloy AL7 under repeated alternating impact bending. The impact toughness and the number of blows to fracture of the specimen were determined as functions of the grain size of the alloy. The corresponding graphs are shown in Fig. 1. The character of the change in the indicated quantities with a change in grain size is almost identical. It should be noted that, in carrying out the tests, the author observed a considerable scatter of the data: for impact toughness, within the limits from 1.85 to 6.31 kgm/cm²; for the number of blows to fracture, from 3840 to 17,970.

Fig. 1. Influence of grain size of the aluminum casting alloy AL7 on the number of repeated blows to fracture (a) and on impact toughness under a single blow (b) [11].

Prof. Ya. B. Fridman [12], referring to the graphs presented by N. F. Lashko, writes: “The final fracture of a specimen in which a fatigue crack has already formed under loading by repeated blows also depends on the magnitude of the impact toughness: the higher the latter, the longer the material will operate with a fatigue crack developing in it.”

Method and results of the investigation

1. Formulation of the problem. The immediate reason for carrying out the investigation was the breakage of rebound-hammer rods during their operation in certain mines and ore—

in mines (Karaganda, Pyatikhátka). Figure 2 shows part of the broken shanks. As can be seen, the cracks originated in the most weakened part of the shank, along the thread, where the holes had been drilled.

Fig. 2. Shanks of jackhammers destroyed during operation.

The results of the analysis of the fractures of the shanks indicate that we are dealing with impact-fatigue failure. Micro-investigations showed that the broken shanks have a coarse-grained structure, corresponding to No. 1–2 on the grain-size scale. The microstructure of the region adjacent to the fracture zone of the shank is shown in Fig. 3.

Under similar loading conditions, not only shanks operate, but also a number of other parts of coal-mining machines. Thus, clarifying the relationship between grain size and the resistance of steel to impact-fatigue failure is a problem of great importance for mining and other branches of machine building.

In addition, an important question is to determine the advisability of carrying out a hardening-and-tempering operation instead of normalizing parts subjected to repeated impact loading.

Fig. 3. Microstructure of the region adjacent to the fracture zone of the shank, ×100.

2. Method and results of the investigation. Standard notched specimens were made from steel 45

Fig. 4. Microstructures of specimens made of steel 45 normalized at temperatures:

a—820°; b—870°; c—900°; d—920°; e—1000°, ×100.

of the Menage type (size \(10 \times 10 \times 55\) mm, notch-root radius \(r = 1\) mm) for determining impact toughness.

In order to obtain different grain sizes, the specimens were normalized from temperatures of 820, 870, 920, and 1000°. In connection with the sharp drop in impact toughness in the temperature interval 870–920°, a control experiment was carried out for a normalizing temperature of 900°. For each temperature point, 3 specimens were taken. The specimens with cast-iron chips were heated in a muffle electric furnace with a holding time of 2 hours. After cooling in air, the specimens were tested on an MK-30 pendulum impact tester. Hardness was measured on a Rockwell apparatus (scale \(H_{R_B}\)). In addition, from each set of three specimens a microsection was prepared for microanalysis and for determining the average grain size. This determination was carried out by the intercept method [13, 14]. Micrographs of specimens normalized at different temperatures are shown in Fig. 4.

Cylindrical specimens \(10 \times 136\) mm were machined from the same steel 45. These specimens (3 pcs. for each temperature point) were normalized according to the same regimes as the specimens for impact-toughness testing. After normalization all specimens were ground to final size. From each set of three specimens a microsection was prepared for checking the microstructure, grain size, and hardness. Tests for repeated-impact bending were carried out on an impact-fatigue machine designed by Prof. A. I. Lampa and Eng. V. N. Polyubin. The scheme for applying impacts and fastening the stationary specimen is shown in Fig. 5. The test results are given in Tables 1 and 2, from which the curves were plotted (Fig. 6).

Table 1

Table 2

3. Discussion of results. In photographs of the fractures of specimens after testing for impact fatigue (Fig. 7), it is evident that the fracture planes have two clearly defined zones: the zone of fatigue-crack development (fine-grained) and the zone of final fracture (coarse-grained). The ratio of the sizes of the zones in specimens with different grain sizes is not the same. In specimens normalized from a temperature of 820°, the zone of fatigue-crack development occupies approximately 50% of the area of the entire section; in specimens normalized from a temperature of 870°, it is somewhat smaller, while in specimens normalized from 1000° this zone amounts to only about 15–18% of the cross-sectional area.

The sizes of the fatigue zones agree well with the results of tests of specimens with similar structures for impact toughness. At one and the same impact energy at which the impact-fatigue tests were carried out, the final fracture of fine-grained specimens, naturally, should occur later than that of coarse-grained specimens, i.e., after a larger number of blows, since fine-grained specimens possess higher impact toughness.

To resist final fracture at a given impact energy, a smaller part of the cross section remaining from the developing fatigue crack is sufficient. This does not mean, of course, that the number of impacts to specimen fracture is always in a single-valued dependence on impact toughness. As can be seen from the graph in Fig. 6, the curves of the dependence of the number of impacts to fracture and of impact toughness on grain size have almost the same character, since the hardness of specimens with different grain sizes remains unchanged. These results confirm the proposition, previously expressed by one of the authors [15], that resistance to impact-fatigue fracture is determined, to a first approximation, by the ratio of two mechanical characteristics of the material: hardness and impact toughness. In the present case the hardness remains constant; consequently, the resistance of the steel to impact-fatigue fracture must depend mainly on the magnitude of the impact toughness.

Fig. 5. Loading scheme for a stationary specimen under repeated impacts in bending.

Fig. 6. Change in impact toughness \(a_k\) and in the number of impacts to fracture of specimens of normalized steel 45 as a function of grain-size change.

Specimens subjected to improvement (Table 2) have a considerably higher resistance to impact-fatigue fracture at the same impact energy, although the impact toughness of these specimens is lower than that of normalized ones. The explanation of this phenomenon follows from the proposition stated earlier: despite the

Fig. 7. Fractures of specimens after testing by repeated impacts. Steel normalizing temperatures:

$a — 820^\circ$; $б — 870^\circ$; $в — 900^\circ$; $г — 920^\circ$; $д — 1000^\circ$; $\times 5$.

a decrease in impact toughness, the higher hardness gives the steel higher resistance under impact-fatigue fracture of the specimen. The fracture of an improved specimen is shown in Fig. 8. The zone of fatigue-crack propagation in this specimen

Fig. 8. Fracture of an improved specimen of steel 45 after testing by repeated impacts. Quenching from 820° in water, tempering at 600°, holding for 30 min., ×5.

is located on two opposite sides and occupies about 70% of the entire cross-sectional area.

CONCLUSIONS

1. The resistance of steel 45 to impact-fatigue fracture decreases significantly with increasing grain size, which makes it necessary to specify the grain size in the steel when manufacturing a number of parts.

2. The resistance of normalized steel 45 to impact-fatigue fracture has approximately the same character of dependence on grain size as does impact toughness.

3. The experimental data confirm the previously advanced proposition that resistance to impact-fatigue fracture depends, to a first approximation, on the ratio of two mechanical characteristics—the hardness and the impact toughness of the steel.

4. Improvement of steel 45 significantly increases its resistance to impact-fatigue fracture in comparison with normalization. Despite some deterioration in machinability by cutting in this case, improvement should be widely recommended as a means of increasing the service life of parts of impact-action machines.

LITERATURE

1. M. M. Tenenbaum. Basic Types of Failure of Parts of Coal Machines, Ugletekhizdat, 1956.

2. M. M. Tenenbaum. Problems of Increasing the Service Life of Parts of Coal Machines. TsITI, Ugletekhizdat, 1956.

3. F. F. Vitman and N. N. Davidenkov. Mechanical analysis of impact brittleness. ZhTF, vol. 7, issue 4, 1937.

1. N. N. Davidenkov. Dynamic Testing of Metals. ONTI, 1936.
2. N. N. Davidenkov. The Problem of Impact in Metallurgy. Academy of Sciences of the USSR, 1938.
3. N. N. Davidenkov. Fatigue of Metals. Academy of Sciences of the Ukrainian SSR, Kiev, 1949.
4. N. F. Bolkhovitinov. Grain Size and the Properties of Steel. Metallurgizdat, 1943.
5. N. N. Davidenkov and E. I. Belyaeva. Investigation of impact fatigue. Metallovedenie i obrabotka metallov, No. 11, 1956.
6. A. I. Lamis. Investigation of the forms of impact-fatigue fractures. Proceedings of the Tula Mechanical Institute, issue 3, 1949.
7. G. I. Pogodin-Alekseev. Increasing resistance to fracture under impact. Vestnik mashinostroeniya, No. 2, 1950.
8. N. F. Lashko. Tests for repeated alternating impact bending. Zavodskaya laboratoriya, No. 5, 1947.
9. Ya. B. Fridman. Mechanical Properties of Metals. 2nd ed., Oborongiz, 1952.
10. M. E. Blanter. Methods for Investigating Metals and Processing Experimental Data. Metallurgizdat, 1952.
11. S. A. Saltykov. Introduction to Stereometric Metallography. Academy of Sciences of the Armenian SSR, Yerevan, 1950.
12. G. V. Toporov. Influence of texture and initial mechanical properties on the impact fatigue of steel 40KhN. Proceedings of the Siberian Physico-Technical Institute at Tomsk State University, issue 34, Tomsk, 1955.

NOTED MISPRINTS

Order 39