Proceedings
of the Tomsk Order of the October Revolution
and Order of the Red Banner of Labor
Polytechnic Institute named after S. M. Kirov

Vol. 224　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　1976

DURABILITY OF HIGH-SPEED STEEL CUTTERS AFTER HIGH-TEMPERATURE TEMPERING

N. S. Kilkov, I. T. Tikhonov, E. P. Filimonova

(Presented by the scientific seminar of the departments
of metal science and technology of heat treatment of metals, technology of metals,
equipment and technology of welding production)

Short-time tempering carried out at elevated temperatures is already finding application in automatic lines for the heat treatment of tools made of high-speed steels. At the same time, there are comparatively few data on the influence of such tempering on the cutting properties of steels [1, 2, 3, 4], as well as on methods for selecting regimes of accelerated tempering and for monitoring them. In this connection, at the suggestion of the Tomsk Cutting Tool Plant, a series of investigations was carried out, the final results of which are summarized in the present article.

In selecting regimes of short-time tempering, one should strive to achieve the highest cutting properties. It is therefore no accident that the question is raised of the optimal tempering regime, and some researchers even conclude that ordinary triple tempering at \(560^\circ\mathrm{C}\) with holding times of 60 min each is not optimal [5]. To avoid possible confusion, it is necessary to dwell on this question in somewhat greater detail.

Indeed, tempering determines the cutting properties of tools not by itself, but only in combination with other factors, the most important of which are the quenching regime and the composition of the steel. Ordinary tempering has a fixed regime and, consequently, cannot be optimal for all cases encountered in the practice of heat-treatment deviations. In the case of optimal tempering, however, its effect should increase, for example, with an increase in the heating temperature for quenching and decrease when it is lowered. In this sense one may agree with the above statement that ordinary tempering is not optimal. On the other hand, long-standing practice in the use of ordinary tempering makes it possible to regard it as quite reliable with respect to obtaining sufficiently good average results in the presence of deviations that are inevitable in the process of heat treatment. Therefore we considered it correct to take as the basis for the investigations precisely triple tempering at \(560^\circ\mathrm{C}\) with holding times of 60 min each.

When comparing the cutting properties of tools after their heat treatment with ordinary and high-temperature tempering, it was first of all necessary to establish a criterion determining the choice of the accelerated tempering regime and making it possible to monitor the correct-

ness of its execution. As such a criterion, it was natural to use magnetic inspection methods, widely employed for checking the results of heat treatment of high-speed steels. In this case, we called high-temperature short-time tempers equivalent to the ordinary one if, after them, the hardened high-speed steel acquired the same magnetic properties as after ordinary tempering [3]. The magnetic properties were monitored on an austenometer specially made for this purpose. The instrument had a conventional differential circuit powered by alternating current and a magnetic-field strength of about 400 Oe. Thus, by equality of the magnetic properties of steels in the case of ordinary and equivalent short-time tempers we meant equality of the austenometer readings in both the first and the second cases. Specimens for monitoring the magnetic properties were cut from calibrated bars and had a diameter of 8 mm and a length of 60 mm. The method for determining equivalent short-time tempers was described in detail in a previously published work [3].

Table 1

As a result of the studies performed, it was shown first of all that equivalent short-time tempers are fully equivalent to ordinary tempering in terms of the hardness and red hardness of the tempered steel [3].

This conclusion was also fully confirmed in the subsequent work, since in all cases in which the austenometer readings were equal, identically hardened specimens after equivalent tempers, but with different tempering temperatures, always had the same hardness.

A comparison of cutting properties after heat treatment with threefold ordinary and short-time tempers was carried out on cutters made of steels of grades R18, R12, R9, R6M3F2, and R9K5. All the cutters whose properties were compared belonged to one batch, were made from one bar of steel, and were quenched simultaneously. Quenching was performed in factory salt baths according to the regimes generally accepted for the indicated steel grades. Tempering was carried out in laboratory bath furnaces with molten nitrate or an aluminum alloy. At the same time, the accelerated tempering had several

variants with small differences in the duration of heating, and cutters whose magnetic properties were the same as the magnetic properties of cutters that had received the usual tempering were selected for testing. The material, heat treatment, and hardness of the cutters selected for testing are given in Table 1. Some difference in the duration of heating during equivalent accelerated tempering of cutters made from identical or similar steels is explained by the fact that their dimensions were changed in several batches. The hardness values given in the table correspond to average values, the deviations from which for a single batch did not exceed 0.5 HRC.

Fig. 1. Dependence of the operating time of cutters to dulling \(T\) on cutter speed \(V\) after heat treatment with different types of tempering: ● — conventional at \(560^\circ\)C; + — threefold high-temperature; × — twofold high-temperature.

Tool-life tests were carried out by turning annealed ShKh15 steel with a hardness that varied, for different blanks, within the range \(207 \div 230\ HB\). All tests whose results were compared were performed on the same work material and with alternating operation of cutters that had undergone conventional and short-time tempering. The cutters operated with a cutting depth of \(2\ \text{mm}\) and a feed of \(0.26\ \text{mm/rev}\). The cutting elements of the cutters had the following geometry: \(\gamma = 17.5^\circ\); \(\alpha = 10^\circ\); \(\varphi = 70^\circ\); \(\varphi' = 10^\circ\). The built-up design made it possible to make a large number of regrindings on one cutter. The results of the tool-life tests are given in the graphs of Fig. 1 in the usual form of the dependence of the cutters’ operating time to dulling on cutting speed. The numbers on the graphs correspond to batch numbers. The difference in the cutting properties of cutters made from one grade of steel is related to the fact that cutters of different designs were used in different batches. In addition, as indicated, the hardness of the material being machined varied somewhat, and the quenching temperatures could differ from the values indicated in the table within the limits customary in plant practice.

In order to obtain at least an approximate quantitative estimate of the compared tool lives of cutters after their conventional and short-

temporary tempers, statistical processing of the results obtained was performed. On logarithmic graphs of enlarged scale, points were plotted according to the test data for each batch. Through these points, first for conventional tempering, a straight line was drawn by the generally accepted method, and, in percentage terms, the ratios of the experimental durability values to its values corresponding to the line at the given cutting speed were determined. Then, if the arithmetic mean of these ratios differed appreciably from 100%, the position of the line was corrected so that the indicated difference did not exceed $\pm 0.2\%$. In a similar way the arithmetic mean was found for the durability after accelerated tempering. Finally, taking the durability values corresponding to the lines as 100% and determining the deviations of the experimental data from them, we obtained the possibility of calculating the mean square errors of the arithmetic means of interest to us.

With the durability of cutters after conventional tempering taken as 100%, the durability of cutters made of steels R18, R12, R9, and R6M3F2 after short-time equivalent tempers for individual batches ranged from 92 to 106%. Moreover, the difference in durability values after conventional and accelerated tempers in all cases did not exceed the limits of the mean square errors of these quantities. For cutters made of steel R18, in total, according to the test results for three batches, the durability of the cutters after short-time tempering was equal to 102%, with a mean square error of $\pm 6.7\%$ and $\pm 6.9\%$ for conventional and short-time tempers, respectively. Evaluating the method as a whole from the test results for cutters of the first six batches, we obtained, with the durability of cutters after conventional tempering taken as 100%, a durability of cutters after short-time tempering equal to 99%, with a mean square error for both types of tempering of $\pm 4.0\%$. Thus, apparently, without risking a large error, it may be considered that the durability of cutters made of the indicated steels after short-time equivalent tempers is maintained at the conventional level within the accuracy of the experiments performed, which is on the order of 5%.

In the case of cutters made of steel R9K5, the durability of cutters after short-time tempering amounted to 80% of their durability after conventional tempering, with mean square errors of $\pm 7.6\%$ and $\pm 9.2\%$.

It is possible that this result was caused by an error in selecting cutters with equivalent tempering regimes, associated with the peculiarities of the change in the magnetic properties of cobalt steels during their tempering. Whereas in hardened high-speed steels without cobalt the saturation magnetization during tempering increases continuously up to a certain limit, in cobalt steels, following the initial increase in saturation magnetization, its decrease is observed [6].

As a result, the readings of the austenometer may not correspond to unambiguously defined tempering regimes, and the question of the possibility of applying the method considered for selecting regimes of equivalent accelerated tempers for cobalt steels should be decided separately.

Conclusions

1. The selection of regimes for high-temperature short-time tempers and monitoring of their execution can be carried out with the aid of austenometers. In this case, short-time tempering may be considered equivalent to conventional tempering if, after both, for identical

samples of hardened steel, equal readings of the instrument will be obtained.

1. The durability of cutters made of steels of grades R18, R12, R9, and R6M3F2 after their heat treatment with three equivalent temperings at 600°C is equivalent to their durability after ordinary tempering.

REFERENCES

1. K. A. Malinina. Short-term tempering of high-speed steel. Metallovedenie i termicheskaya obrabotka metallov, No. 3, p. 47, 1960.

2. P. A. Yudkovskii, V. Ya. Bulanov, Yu. N. Zhuravkov, A. P. Shevel. The influence of heat treatment on the durability of drills. Stanki i instrument, No. 12, p. 47, 1963.

3. I. T. Tikhonov, N. S. Kilkov, E. P. Filimonova. Short-term tempering of high-speed steel R18. Izv. vuzov. Chernaya metallurgiya, No. 10, p. 112, 1964.

4. N. S. Kilkov, I. T. Tikhonov, E. P. Filimonova. The influence of short-term temperings on the properties of high-speed steel. Izv. TPI, Vol. 147, p. 206, 1966.

5. P. A. Andzhos. Study of the tempering of high-speed steel. Proceedings of the scientific-technical conference dedicated to the tenth anniversary of the Vilnius Drill Plant. Vilnius, 1967.

6. Yu. A. Geller. Tool Steels. Metallurgizdat, Moscow, 1964.