ON THE QUESTION OF COMPARATIVE BEND TESTING OF CAST PRECIPITATION-HARDENING TOOL ALLOYS

A. D. KLEMENTYEV

(Presented by a scientific seminar of the departments of physical metallurgy, equipment and technology of heat treatment of metals, technology of metals; equipment and technology of welding production)

In the first publications [1–3] of the results of our studies of iron–cobalt–tungsten–molybdenum alloys, the expediency of using them and the prospects for their further development as tool materials were demonstrated.

In subsequent studies of these alloys by other authors, various technological variants and alloy compositions were used. In this connection, it becomes necessary to compare the results obtained by different researchers, and also to clarify the test conditions and the states of the alloys for which such comparisons are most appropriate. This article considers the special features of carrying out bend tests of the alloys under study.

In the publications cited above [1–3], the experimental material is presented in a very condensed form, and the conditions for carrying out the bend tests are not everywhere shown with sufficient clarity; this in some cases [4] makes it difficult to obtain results that are sufficiently comparable.

Bend tests are a very informative and convenient characteristic of the mechanical properties of cast precipitation-hardening cutting alloys. The concentration of stresses in the well-controlled surface layers of a specimen during bending promotes the detection of an alloy’s tendency toward various kinds of brittleness, including brittleness that arises both in the process of precipitation hardening and as a result of defects in the cast structure, which are also revealed in the fracture of the specimen. In addition, this method is of special value in testing precisely cutting alloys, owing to the known analogy with the conditions of the stressed state of the cutting part of a tool.

As studies [1–3] have shown, in alloys with high bending strength the wear process during cutting proceeds gradually and regularly, and with sufficiently high hardness and red hardness, good durability of the alloys is achieved. In alloys in whose tests periodic chipping of the tool cutting edge occurs, accompanied by a sharp decrease in durability, the bending strength is not high.

Bending strength has proved to be a good structurally sensitive characteristic. Alloys possessing good bending strength and durability in cutting are distinguished by a homogeneous, fine-grained microstructure and fracture structure.

Pursuing only the aim of comparing alloys by their bending strength and of making rational use of small melt volumes and specimen sizes,

castings ($10—15 \times 25 \times 100$ mm), nonstandard test conditions were used on small specimens $2.5 \times 6 \times 10$ mm, with a distance of 7 mm between supports. In addition, on small specimens the various types of heterogeneity of structure and properties, which depend on the dimensions of the castings, have less effect [5].

The relatively large width of the specimen in comparison with its height was chosen in order to smooth out macro- and microscopic heterogeneity in the cross section of the specimen (many alloys are coarse-grained; nonuniform precipitation of the excess phase occurs, etc.) and to average the effect of various small stress concentrators. This provides greater stability of the specimen and a more uniform application of the load during testing.

Fig. 1. General view of the apparatus for determining bending strength

Fig. 2. Apparatus for determining bending strength in disassembled form

When the bending force is located close to the supports and relatively high above them, it becomes less concentrated, since in fact it acts over a certain middle portion of the specimen. A complex state of stress arises, in which, in addition to bending, compression of the specimen begins to play a significant role.

As a result, the bending strength should be expressed by higher values than in standard tests. It should be noted that the adopted deformation scheme corresponds more closely to the operating conditions of the cutting part of the tool than do standard bend tests.

The tests were carried out on a specially manufactured apparatus (Figs. 1 and 2). The specimen is placed in the longitudinal groove of a self-aligning swivel cradle. This ensures uniform contact of the prism edge with the specimen surface at the moment the load is applied. The position of the specimen relative to the supports (above the opening in the cradle) is aligned by means of lateral stops adjusted by screws at the ends of the cradle. The position of the prism edge relative to the supports (and, consequently, relative to the specimen) is fixed in the axial direction of the cradle by a pin entering the annular groove of the cradle. The faces of the prism form an angle of $45^\circ$. The prism edge is blunted (rounding radius 0.5 mm). The fixture is mounted on a mechanical or hydraulic press that permits the load acting on the prism to be recorded.

The bending strength, taking into account the brittle nature of the fracture, was determined by the formula

\[ \sigma_{\text{bend}}=\frac{3Pl}{2bh^2}\ \text{kg/mm}^2, \]

where

\[ \begin{aligned} P&\text{— breaking load, kg;}\\ l&\text{— distance between supports, mm;}\\ b&\text{— specimen width, mm;}\\ h&\text{— specimen height, mm.} \end{aligned} \]

As was shown in [5], in many cases the alloys differed in the nonuniformity of structure and properties across the section of the castings, which must be taken into account when choosing the location for cutting specimens from the castings. The latter, in turn, is related to the conditions under which the casting is formed. The molten metal was poured into the mold so that the stream entered its central part. The casting forms with its maximum section lying in the horizontal plane. This produces symmetry in the structure and properties of the casting. The best-quality regions of the castings prove to be the areas most distant from the pouring point (mainly owing to the higher cooling rate).

To eliminate random results, tests were usually carried out on two specimens taken from one location in the casting. When a large number of tests had to be performed, for example under different heat-treatment variants, specimens were taken from symmetrical regions of the casting. When comparing the properties of different alloys, specimens from corresponding regions of different castings were used.

Of great importance is the choice of the state of the alloy for which determination of strength is most expedient. This is especially important in comparative tests of a large number of alloys, with a minimum volume of research operations, in order to obtain the quickest and most correct assessment of the alloys. From the standpoint of serviceability, it would be expedient to compare alloys in their working state, i.e., after the complete cycle of their treatment has been carried out. However, as studies [1–3, 5] have shown, in this case the strength depends strongly on the slightest variation in heat treatment, while the most advantageous regimes are still insufficiently clarified and, for different alloy compositions, may differ considerably.

Under these conditions it is expedient to evaluate the strength of alloys in their initial as-cast state, since it substantially determines the strength after various treatment variants as well. Studying the effect of different types of heat treatment on strength makes it possible to assess qualitatively the possible change in strength as a function of the structure and composition of the alloys.

It is of interest to compare the results of bend tests of alloys in our investigations and in work [4], in which similar cast precipitation-hardening alloys are studied. In contrast to our conditions, in that work the bending strength was determined for heat-treated alloys on specimens of somewhat larger dimensions in length and height ($3 \times 3 \times 15$ mm).

We carried out comparative bend tests of specimens $2.5 \times 6 \times 10$ mm with a distance between supports $l = 7$ mm and of specimens $3 \times 3 \times 15$ mm with corresponding distances of 7 and 12 mm. The materials tested were high-speed steel R18 (from a bar 10 mm in diameter), heat treated according to the usual regime to a hardness of 63 HRC, and a precipitation-hardening alloy (20% W, 20% Co, 1% Mn, 1% Ni, 0.2% Ti, 0.5% V) in the initial state (42 HRC) and after tempering at 670°C for 30 min (65 HRC).

Under the indicated test conditions, the following values of bending strength ($\sigma_{\text{bend}}$, kg/mm²) were obtained:

Attention is drawn to the closeness of the strength values obtained by the compared methods on specimens of high-speed steel (\(373\) and \(361\ \mathrm{kg/mm^2}\)), which follows from the mutually compensating influence of height and length and from the features of the stressed state of the specimens. However, such agreement is not observed in tests on the precipitation-hardening alloy. In this case, the lower strength of the longer specimens (\(319\) compared with \(383\ \mathrm{kg/mm^2}\)) is explained by the small size of the castings and by the more pronounced structural inhomogeneity of the longer specimens compared with the short specimens adopted in our studies. Heat treatment of the precipitation-hardening alloy, by intensifying the inhomogeneity of properties owing to its structural inhomogeneity, further increases the difference in the strength values of the two types of specimens (\(216\) and \(272\ \mathrm{kg/mm^2}\)). The specimens of high-speed steel, by contrast, were distinguished by a very high degree of structural homogeneity.

As was to be expected, changing the conditions of the stressed state by reducing the distance between supports (with the same specimen cross section) leads to an increase in the values of bending strength (for example, for R18 steel, \(361\) and \(432\ \mathrm{kg/mm^2}\)). This difference increases for the precipitation-hardening alloy, and especially in the heat-treated state. It should be noted that there is an increased scatter of strength values for the precipitation-hardening alloy in tests with a large distance between supports, and especially for heat-treated specimens. The use of \(2.5\times6\times10\) mm specimens gives more stable results under our research conditions.

Thus, when making comparisons of the alloys studied in terms of bending strength, it is necessary to take into account the specific nature both of the test conditions and of the alloys themselves. As for the alloys investigated in [4], our alloys corresponding to them in composition are at approximately the same strength level.

Conclusions

1. Bending strength is a good structure-sensitive characteristic of the properties of cast precipitation-hardening alloys.

2. Determination of the bending strength of the alloys studied was carried out in [1–3] on small specimens. The strength values obtained in this way prove to be overestimated in comparison with those obtained in standard tests.

3. In comparative tests of a large number of alloys, bending strength was determined in their initial as-cast state [1–3], which made it possible to identify, most simply and rapidly, alloys promising for further study.

References

1. A. D. Klement’ev. Study of cast precipitation-hardening cutting alloys. Izvestiya vuzov MVO SSSR. Chernaya metallurgiya, No. 3, publ. Siberian Metallurgical Institute, 1959.

2. A. D. Klement’ev. Cast precipitation-hardening cutting alloys. Proceedings of the Conference of Toolmakers of Western Siberia, collection 1, Moscow, TsINTI, 1962.

3. A. D. Klement’ev. Cast precipitation-hardening cutting alloys. Abstract of Candidate’s dissertation, Tomsk State University Publishing House, Tomsk, 1962.

4. Yu. A. Geller et al. Cast precipitation-hardening alloys with high hardness and strength. Metallovedenie i termicheskaya obrabotka metallov, No. 8, 1968.

5. A. D. Klement’ev. Study of the nonuniformity of properties across the section of castings from iron–cobalt–tungsten–molybdenum alloys. Izvestiya TPI, vol. 173, Tomsk, 1970.