Low-temperature strength of thermally hardened rolled rounds

LOW-TEMPERATURE STRENGTH OF HARDENED ROLLED ROUNDS K. F. S tarodubov , Yu. P . Gu l ' , and A. S. Gu levsk i i THERMALLY UDC 669.04:621.771 It is known that thermal hardening of roiled products of low-carbon steel carr ied out under optimal conditions increases resistance to brittle fracture in addition to increasing resistant to elastic and plastic deformation. Resistance to brittle fracture is expressed, in particular, by a corresponding decrease of the temperature of cold britt leness in comparison with the hot-rol led or normalized condition [1]. The data on an increase of cold resistance of low-carbon steel as a result of its thermal hardening have been obtained mainly from the results of serial dynamic bending tests. However, these results are difficult to use directly in engineering calculations in which values of the allowable stresses are used. In addition, an evaluation of the cold resistance of steel based on the results of its dynamic bending tests is too stringent for the structures and parts of machines operating mainly under static load conditions [2], which leads to an unsubstantiated overstatement of the necessary temperature margin of toughness. At the same time, there is comparatively little information on the low -temperature strength of ther- mally hardened low-carbon steel determined, for example, in static tension [3, 4]. There are practical ly no data obtained on ful l -scale steel specimens subjected to thermal hardening from heating during rolling under industrial conditions. The present investigation was carr ied out on 14-, 16-, and 20-mm-diam. rolled rounds subject to thermal hardening from heating during rolling on the experimental industrial unit of the Makeevka Metal- lurgical Plant. These rounds were taken also in hot-rol led and normalized conditions for comparison. The chemical composition of the steel grades investigated and the main parameters of the heat- treatment conditions are presented in Table 1. TABLE 1 Steel grade St. 3sp 0,17 St. 5sp 0,33 Chemicai composition, % by mass C Mn Si 0,53 0,21 0,61 0,27 0,043 0.038 Normalization from furnace heating P heatingtem- perature, ~ 0,021 860 0,015 860 holding time, rain 30 3O Steel grade St. 3sp St. 5sp from electrical heating' heating tern- heating perature, ~ time, rain 960 5 960 3 Thermal hardening from heating during rolling I variant: thermal hardening with self ~ II variant: thermal hardneing + fur- III variant: thermal hardening + elec- tempering Inace tempering tric ~em~ hardening cooling time self-temper-thardening eoolingtime'tempering thardening 'eooling timettempering tempera- in quenching ingtempera-ltempera- in quenching temperature~tempera- [n quenching~emperature, ture bath, sec ture, ~ Iture ,bath, see ~ [ture bath, sec t~ 15 i0 100 350 900 860 20 20 350 350 900 860 20 20 450 450 900 86O Dnepropetrovsk Metallurgical Institute. Translated from Problemy Prochnosti , No. 8, pp. 103-107, August, 1973. Original article submitted February 14, 1972. �9 1974 Consultants Bureau, a division of Plenum Publishing Corporation, 227 g/est 17th Street, New York, N. Y. 10011. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission of the publisher. A copy of this article is available from the publisher for $15.00. 1007 'o, kg/: ~n 2 " St. 3sp 50 -140 -lo0 60 -20.20 -140 -100 -60 -20 t,oc -I~0 -]00 -60 -20 t20 -]~0 -100-6# -2o L,~s a b c d Fig. 1. Mechanical properties in static tension of full-scale specimens of 20 mmdiam, rounds of steels St. 3sp and St. 5sp as a function of the test temper- ature in a hot-rol led condition (a, e) and thermally hardened according to the I variant (b, d): 1) crt; 2) Cry; 3) cry. The mechanical properties in static tension were determined on smooth and notched full-scale speci- mens of the aforementioned 350-500 mm long rounds in the temperature range from +20 to -140 ~ on a UMM-50 machine at an average strain rate of 4" 10 -3 sec -i with recording of the stress - strain diagram. The notch was a c ircular groove with a depth from 2 to 5 ram, 0.25 mm radius of curvature of the bottom, and 60 ~ angle of opening, Owing to the large size of the investigated specimens, they were tested at negative temperature by the following method. The specimens were placed in a precooling chamber and held in a liquid coolant (mixture of liquid nitrogen with hydrolytic alcohol or petroleum ether) for the necessary time with some super- cooling as opposed to the prescribed test temperature for compensating the heating of the specimens during transfer and installation in the clamps of the test machine. During the test the specimen was placed in an insulating device - a stainless steel hollow cylinder filled with liquid nitrogen. At the end of the test the temperature of the specimen was measured immediately by means of a contact thermometer. A shortcoming of this method is that a certain temperature gradient is obtained over the length and section of the test specimens. Special experiments showed, however, that this gradient is small: for specimens with a length of 500 mm and diameter of 20 mm it was up to 5 ~ over the section and up to 2 ~ over the length. The microstructure of the investigated steels was studied under an optical microscope and electron microscope by means of carbon replicas; the fine structure of the fractures of the ruptured notched spec- imens in the region of tough (test temperature +20 ~ and brittle (-140 ~ fracture was studied also by elec- tron microscope by means of carbon replicas.* It follows from the data of Table 2, which gives the average values of the mechanical properties of the investigated steels at room temperature after various heat treatment, that thermal hardening of steel St. 3sp increased its strength character ist ics in comparison with the hot-rol led and normalized condition by about twofold and of steel St. 5sp by 1.7-2 times. Here different variants of hardening provided approxi- mately the same levels of strength for each of the investigated steel grades, which corresponded to the IV class for steel St. 3sp and to the V class of strength for steel St. 5sp according to GOST 10884-64. With a decrease of the temperature of testing smooth specimens of the investigated steel grades there occurred the usual change of character ist ics of strength and plasticity: the former increased and the latter accordingly decreased (Fig. 1). However, heat treatment, mainly thermal hardening, had a subs~n- tim effect on the quantitative character of the change of properties with a decrease of temperature. A rather marked drop of the yield strength and nominal tensile strength, byabout 40 kg/mm 2 (Fig. la), was observed for steel St. 3sp in hot-rol led and normalized condit.ions at temperatures below -100 to *The electron microscopic investigation was carr ied out together with A. I. Karnaukh and N. V. Kolpov- skaya. 1008 P, kg 25000 ZOO00 5000 i ,/Y / ]0 20 F ig . 2 Q 50 40M F ig . 2" Techn ica l s t ress - s t ra in d iagrams of smooth fu l l - sca le spec imens of 20 mm d iam. rounds of s tee l St. 3sp at temperatures +20 ~ (1), -120 ~ (2), and - 14 0 ~ (3). F ig . 3. F ine s t ructure of f rac tures of fu l l - sca le notched spec imens of s tee l St. 5sp that fa i led at d i f ferent temperatures in a hot - ro l led condit ion (a, c) and thermal ly hardened by the I var iant (b, d) (5000x), TABLE 2 ~ 2 rm~", kg % Heat treatment d'm2 %, 4, % 40 62 45 6O 47 6O 3O 35 3O 32 32 32 3O 65 65 65 80 64 I0 45 1o---6 ~-o ~ u 80 66 18 55 io--~ ~ -~ ~-o 80 62 15 55 10--0 ~ ~ 4-2 Hot-rolied condition Normalization from furnace heating from e leetrica 1 heating Thermal heating I Variant II variant III variant -120 ~ . Ev ident ly , th is is re la ted with premature br i t t le f rac ture , which is conf i rmed, in par t i cu la r , by the tens i le s t ress - s t ra in d iagrams (Fig. 2). These character i s t i cs do not drop for thermal ly hardened steel St. 3sp in the in - ves t igated temperature range (F ig. lb) . The nominal rupture s t ress , determined on notched spec imens , for al l the invest igated heat - t reatment condi - t ions depends nonmonoton ica l ly on the test temperature : with a decrease of temperature it at f i r s t inc reases , and then drops ra ther sharp ly . Since in this case the va lues of the nominal rupture s t ress a re lower than the va lues of the tens i le and y ie ld s t rengths determined on smooth spec i - mens , the ind icated drop can be assoc ia ted with t rans i t ion into the br i t t le f rac ture reg ion , and the temperature of the s tar t of the drop of the nominal rupture s t ress can be taken as the temperature of co ld br i t t leness for notched spec i - mens (Tcb). In the tough f rac ture reg ion the nominal rupture s t ress on notched spec imens is h igher than the tens i le Note. The data for steel St. 3sp are given in the m~mera- tor and for steel St. 5sp in the denominator. s t rength on smooth spec imens . Th is occurs due to a decrease of the ra t io of the max imum tangent ia l s t resses to the max imum normal s t resses as a consequence of the more dras t i c state of s t ress in the re - gion of the notch. In the br i t t le f rac ture reg ion, where the process is cont ro l led main ly by the magnitude of the max imum normal s t resses , the p ic ture , natura l ly , changes to the oppos i te . The t rans i t ion to the br i t t le f rac ture reg ion upon a drop o f the nominal rupture s t rength is conf i rmed a lso by the change of char - ac ter of the f rac ture . The main data on the ef fect of the heat t reatment condi t ions and notch depth on the temperature of co ld br i t t leness are given in Table 3, f rom which fol lows that in stat ic tens ion thermal harden ing by al l var iants for both s tee l g rades reduces the temperature by 20-40 ~ An increase of the notch depth f rom 3 to 5 mm for s tee l St. 3sp does not have a substant ia l ef fect on the temperature of co ld br i t t leness Tcb , where - as an increase of notch depth f rom 2 to 5 m for s tee l St. 5sp main ly inc reases Tcb in the thermal ly hard - ened condi t ion and condi t ion normal i zed f rom furnace heat ing. Thermal harden ing of the invest igated s tee ls inc reased the nominal b reak ing s t ress in compar i son with the hot - ro l led or normal i zed condi t ion at the lowest invest igated temperature ( -140 ~ by 1 .7 -2 t imes 1009 TABLE 3 Steel grade St. 3sp St. 5sp Notch depth, mi l l hot-rolled condition -80 -80 -80 --80 Temperature of cold brittleness Tcb, ~ normalization thermal ha g from fur- from elec- nace heat- tricalheat- ing -80 -80 --80 -80 I variant II varL --I00 --i00 --120 --i00 -80 -80 - I00 -80 ing - i00 --i00 - i00 - - 120 III varia~t .I0O -12v - I00 TABLE4 Steel grade St. 3sp St, 5sp hot-rolled condition Notch depth 55 5 58 2 70 5 70 Nominal rupture stress, kg/mm z normalization from fur- from elec- nace heat- tricalheat- ing Img 59 54 56 52 70 65 67 69 thermal hardening I variant II van 92 85 90 92 150 145 130 140 III variant !03 105 148 145 (Table 4). At this temperature different notch depths and thermal hardening conditions did not have a sub- stantial effect on rupture stress. A comparison of the levels of normal rupture stresses at temperatures + 20 ~ and - 140 ~ i.e., in the brittle and tough fracture region, showed that for steel St. 3sp in a hot-rol led and normalized condition the rupture stress at -140 ~ is about 15 kg/mm 2 less than at +20 ~ Thermal hardening of steel St.3sp reduces this difference, and for thermally hardened steel St. 5sp the rupture stress at -140 ~ is now about 15 kg /ram 2 greater than that at room temperature. Thus, for a comparatively reliable evaluation of the low-temperature strength of thermally hardened steel it is possible to use the character ist ics of the rupture strength determined at room temperature. The values of the low-temperature strength determined on thermally hardened rounds of steels St. 3sp and St. 5sp are comparable to the corresponding character ist ics of the mostcommon grades of low-allow steel' grades [3-4]. Aeomparison of the data obtained permits the conclusion that the increase of the low-temperature strength of rounds of low-carbon steel as a result of thermal hardening in comparison with the correspond- ing character ist ics in a hot-rol led or normalized condition is related with an increase of resistance to brittle fracture. This agrees with the intensity of the temperature dependence of the yield strength after various heat treatment and with the results of the fraetographic investigation. The intensity of the temperature dependence of the yield strength of thermally hardened steel de- creases considerably in comparison with such steel in a hot-rol led or normalized condition (see Fig.!l . At the same time, weakening of the temperature dependence of the yield strength can be related with a de- crease of the propensity of the metal for cold brittleness [5-7]. In the tough fracture region the fractures of the investigated steels have a characterist ic ?'cup" struc-- ture, along with which "tough" cleavage surfaces are sometimes found (Fig. 3a, b). The heat treatment conditions used had mainly a quantitative effect on the character ist ics of the fine structure of the fracture in the case of tough fracture. Statistical treatment of the results of the fractographic investigation showed i010 Fig. 4. Microstructure of steel St. 5sp in a hot-rol led (a) and thermally hardened (b) condition. that the average size of the "cup" of thermally hardened steel is two-three times less than the size of the cups of steel in a hot-rol led or normalized condition. To some extent this agrees with the greater dis- persion of the carbide particles and their more uniform distribution in the matrix after thermal hardening in comparison with that in the hot-rol led or normalized state (Fig. 4). In the case of a macrobritt le structure of the fractures their fine structure can be different depend- ing on the heat treatment conditions. For the hot-rol led or normalized condition the fine structure of the fractures of the investigated steels is character ized mainly by the presence of "open" planes of sufficient length and a "r iver" pattern (see Fig. 3c). For thermally hardened steel areas of tough cleavage and with a cup structure are found along with areas of the fracture surface shown in Fig. 3c. Thus thermally hardened steel retains the capacity for plastic relaxation of elastic st resses at the lowest temperature considered here, which apparently is one of the main causes of the high low-tempera- ture strength of this steel in comparison with the low-temperature strength of hot-rol led or normalized steel. 1o 2. 3. 4. 5. 6. 7. L ITERATURE C ITED K. F. Starodubov et al., Thermal Hardening of Rolled Products [in Russian], Metallurglya, Moscow (1970). V. S. Ivanova et al., Fatigue and Brittleness of Metallic Materials [in Russian], Metallurgiya, Mos - cow (1968). S. I. Gudkov, Mechanical Properties of Steel at Low Temperatures [in Russian], Metallurgiya, Mos - cow (1967). P. F. Koshelev and S. E. Belyaev, Low-Temperature Strength and Plasticity of Structural Materials [in Russian], Metallurgiya, Moscow (1967). J. P. Lowe, in: Fracture of Solids [Russian translation], Metallurgiya, Moscow (1967). Yu. P. Gul', Yu. A. Krishtal, and V. S. Siukhina, in- Interaction between Dislocations and Impurity Atoms and Alloys [in Russian], Izd. TPI, Tula (1969). Yu. P. Gul' and V. S. Siukhina, "Thermal hardening of rolled product," in: Ferrous Metallurgy Papers of the USSR Ministry of Ferrous Metallurgy [in Russian], No. 30, Metallurglya, Moscow (1969). i011