STUDY OF THE CUTTING PROCESS OF BERYLLIUM BRONZE

M. F. POLETIKA

(Presented by Prof., Doctor A. M. Rozenberg)

One of the main reasons hindering the creation of a scientifically substantiated theory of the machinability of metals by cutting is the insufficient development of the theory of chip formation, especially that part of it which concerns the influence of the properties of the material being machined—in particular, its mechanical properties—on the process of chip formation.

An experimental study of the influence of the mechanical properties of the material being machined on the cutting process can be carried out in two ways. The first consists in conducting experiments with a large number of different machined materials and subsequently performing a painstaking analysis of their results. Quite apart from the very great laboriousness of such a research method, it may also fail to achieve its aim, since different machined materials differ not only in their mechanical properties but also in many other properties. It is rather difficult to isolate the influence on the cutting process of each group of properties separately.

The second way consists in artificially changing one or another set of properties of the metal being machined. With respect to mechanical properties, this is best achieved by carrying out preliminary heat treatment of the metal under investigation under different conditions. Such a research method leads to the goal considerably more quickly than the first.

Having set ourselves the task of investigating the influence of the mechanical properties of a material on the cutting process, we chose, as the material to be machined, BrB2 bronze containing 1.87% beryllium. This particular alloy attracted us because, when the heat-treatment conditions are changed, it changes its mechanical properties but does not change its microstructure. This makes it possible to isolate the influence of mechanical properties on the cutting process in a “pure” form.

We had beryllium bronze available only in the form of bars 24 mm in diameter. Therefore, specimens 100 mm long each were cut from the bars. All specimens were subjected to quenching, consisting of heating to 780° followed by cooling in water. After this they were sorted into 4 groups and tempered. By varying the temperature and duration of tempering for each of the groups, we obtained four different hardnesses: HB = 111, 200, 321, 360.

The sharpening geometry of the cutter remained constant in all experiments: rake angle—10°, clearance angle—10°, angle of inclination of the main cutti-

cutting-edge angle—0°, plan angle—70°. The cutting speed was varied over the entire range that could be provided on the 1K62 lathe with the limited diameter of the specimen. The experiments were carried out with four feeds: 0.07; 0.15; 0.26 and 0.34 mm/rev. The depth of cut in all experiments remained constant and equal to 2 mm.

In each experiment, the chip compression, the length of chip contact with the tool, the temperature, and the cutting force were measured. To measure the cutting temperature, the natural-thermocouple method was used; the calibration of the latter was performed in the usual way, but separately for each group of specimens.

The cutting force was measured by means of a two-component elastic-electric dynamometer with inductive transducers. The readings of the dynamometer were recorded by three microammeters arranged in a row on a panel together with the instrument for measuring the cutting temperature. During the experiment, after the readings of all instruments had stabilized, the panel was photographed with a camera fixed on the stand. In this way the simultaneity of all four readings was ensured.

In processing the measurement results, it was necessary to separate, from the total forces measured by the dynamometer, the forces acting on the rake face of the tool. For this purpose we used the method of extrapolating to zero feed the lines of dependence of the cutting force on feed under conditions of constant chip compression.

The experiments showed that the mechanical properties of the work material, in particular its hardness, have a very substantial influence on the entire process of chip formation. This influence is noticeable even in direct observation of the cutting process, but is revealed most clearly in visual examination of the chip and in measuring its compression.

Curves of the variation of chip compression as a function of cutting speed for one of the feeds are given in Fig. 1. From them it follows first of all that, as the hardness of the work material increases, chip compression continuously decreases. The process of chip formation in cutting soft bronze is very similar to the process of chip formation in machining low-carbon steel. Here the chip is continuous, and the compression is large. In the region of low cutting speeds a pronounced built-up edge is observed on the tool and, as a result of built-up-edge formation, the curve of the dependence of chip compression on cutting speed has a characteristic hump and trough.

Unlike steel, however, the drop in compression after the speed passes into the built-up-edge zone occurs very sharply. This is apparently explained by the fact that a highly stable built-up edge arises at once, which within a certain range of speeds changes its stability little, and with a further increase in cutting speed disappears as quickly as it appeared.

With increasing hardness of the bronze up to a certain limit, the form of the curve \(\zeta=f(v)\) is preserved, but the chip compression continuously decreases, including on the hump of the characteristic curve of its variation. The only exception is the zone of built-up-edge formation, where the compression remains approximately constant.

With a further increase in the hardness of the work material the built-up edge disappears, and with increasing speed we observe only a smooth decrease in compression. The chip gradually changes its form, passing from continuous to serrated (the elements are only indicated), and then to elemental. In this case the serrated and elemental chips have a characteristic structure: there is no texture within the element, and only a more or less thin contact layer is visible, as a rule strongly textured, which binds the elements together.

Very often such a chip (this is especially true of segmented chips) is, in appearance, difficult to distinguish from a continuous chip. The boundaries of the elements are revealed only after polishing and etching its side surface. It should be noted that a chip of the same structure is obtained when machining titanium alloys.

Fig. 1.

The decrease in chip shrinkage with increasing cutting speed in the region where built-up edge is absent is accompanied, as is seen in Fig. 2, by a decrease in the average coefficient of friction of the chip against the tool; from Fig. 3, which gives the values of the average normal \(q_N\) and average tangential \(q_F\) stresses on the rake face, calculated from the apparent contact area, we are also convinced that the decrease in the coefficient of friction is due to an increase in the average normal stress. The specific friction force, however, is practically independent of cutting speed. This means that the mechanism by which cutting speed affects the chip-formation process when cutting BrB2 bronze is the same as when machining steel, for analogous results were obtained by N. N. Zorev [1] in turning steels.

From consideration of the same Figs. 2 and 3 it follows that the hardness of the material being machined has essentially the same effect on the coefficient of friction as does an increase in cutting speed. With increasing hardness the coefficient of friction decreases, and this again is the result of an increase in the average normal pressure on the rake face of the tool.

The circumstance that any change in chip shrinkage is associated with a change in the average coefficient of friction of the chip against the tool is a fact long since established by researchers. However, most of them are inclined to regard the coefficient of friction as a certain physically independent quantity which ultimately determines shrinkage. The results of the present investigation convince us that this is not so.

Fig. 2.

Graph labels: \(\mu\); \(v\), m/min.

Legend: Hardness HB: \( \times \)—111; \(\Delta\)—200; \(\bullet\)—321.

Fig. 3.

Graph labels: \(q_N\), kg/mm\(^2\); \(q_F\), kg/mm\(^2\); \(v\), m/min.

Legend: Hardness HB: \( \times \)—111; \(\Delta\)—200; \(\bullet\)—321.

In fact, the coefficient of friction under cutting conditions, when friction is not necessarily accompanied by deep surface deformations in the contact layers of the chip, as a certain independent physical quantity, should be determined by the resistance to deformation in these layers, that is, by the magnitude of the average shear stress in the zone of contact between the chip and the cutter. However, the disagreement of this quite natural assumption with experiment follows both from our experiments and from those of N. N. Zorev, since as the cutting speed increases the shear stress on the rake face remains constant, while the coefficient of friction decreases.

If in the given case some doubts might still arise, since an increase in cutting speed entails an increase in temperature, which changes the conditions of friction, they should now disappear. With an increase in the strength characteristics of the metal being machined, one should expect an increase in the specific friction force, and it indeed increases, as we see from Fig. 3. The coefficient of friction, however, does not increase in this case but, on the contrary, decreases. The cutting temperature in the given case plays no role, since it can be shown on the basis of our own experiments that the indicated tendency does not depend on it.

Consequently, the coefficient of friction in cutting is not an independent physical parameter of the material being machined. Apparently, it represents a certain generalized characteristic of the stressed state in the cutting zone and, of course, must vary in accordance with the chip shrinkage, which is a measure (albeit not a fully precise one) of the deformation in this zone.

In this connection, it is inappropriate to pose the question of which is determinative: the coefficient of friction or the chip shrinkage. Both of these characteristics of the cutting process are conditional and are functions of independent parameters, which include the geometry of the cutter and the elements of the cutting regime, as well as certain characteristics of the material being machined and of the tool material.

REFERENCES

1. N. N. Zorev. Questions of the Mechanics of the Cutting Process. Mashgiz, 1956.