Sie sind auf Seite 1von 15

M. Wendel et al.

: Bearing steels for induction hardening

M. Wendel, F. Hoffmann, W. Datchary

Bearing steels for induction hardening – Part I


Wälzlagerstähle für die Induktionshärtung – Teil 1

Abstract/Kurzfassung
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

Surface induction hardening, through induction hardening, Randschichthärten, Durchhärten und Anlassen von Wälzla-
and induction tempering of components like rolling elements gerkomponenten mittels induktiver Erwärmung sind etablier-
or rings are established in production and are further devel- te Produktionsprozesse, welche im Bereich der Forschung und
oped in research and development. Among others, the used Entwicklung weiterentwickelt werden. Neben vielen Faktoren
steel grade for induction hardening is one important factor hat die Werkstoffauswahl einen maßgeblichen Einfluss auf das
affecting the induction hardening result. The present study Wärmebehandlungsergebnis. In der vorliegenden Arbeit wird
benchmarks potential and existing bearing steels in terms of eine Auswahl üblicher sowie für das Induktionshärten poten-
their induction hardening response. Steel grades with low to ziell interessanter Wälzlagerstähle bezüglich ihres Verhaltens
high carbon contents and different alloying strategies were beim induktiven Härten bewertet. Es wurden Stähle mittlerer
selected. The robustness of the heat treatment result depend- bis hoher Kohlenstoffgehalte mit unterschiedlichen Legie-
ing on the selected steels and their prior conditions, prior rungsstrategien ausgewählt. In Abhängigkeit der verschiede-
to induction hardening, shall be pointed out. A dilatometer nen Stähle sowie verschiedener Gefügezustände vor dem Här-
study was used to compare the sensitivity of martensite start ten wurde die Stabilität des Härteergebnisses erörtert. Dabei
For personal use only.

temperature (MS), hardness after quenching, prior austenite wurde mithilfe einer Dilatometerstudie die Veränderung der
grain size, and appearance of non-martensitic transformation Martensitstarttemperatur, der Härte, der ehemaligen Austenit-
products, for varied process parameters. The study showed the korngröße und der nicht martensitischen Umwandlungsbe-
pronounced effect of the steels` prior microstructure, prior to standteile des Gefüges für veränderte Austenitisierbedingun-
induction hardening, on the hardening response. The alloying gen verglichen. Der Gefügezustand vor dem Härten zeigte sich
with silicon and manganese reduced the amount of non-mar- als wichtige Einflussgröße für das Härteergebnis. Das Legieren
tensitic transformation products after hardening. Increasing mit Silizium und Mangan führte zu einer Reduzierung der
amounts of alloying elements led in general to larger changes nicht martensitischen Umwandlungsbestandteile des Gefüges
in MS temperature and lower hardness after hardening. n nach dem Härten. Allgemein führten höhere Gehalte an Le-
Keywords: Induction hardening, bearing steels, 100Cr6, 50CrMo4, Ovako 832 gierungselementen zu größeren Veränderungen der Marten-
sitstarttemperatur mit veränderten Austenitisierbedingungen.
Mit zunehmendem Legierungsgehalt wurde zudem eine ge-
ringere erreichbare Härte gemessen. n
Schlüsselwörter: Induktionshärten, Wälzlagerstahl, 100Cr6, 50CrMo4, Ovako 832

Authors/Autoren: Michael Wendel, AB SKF, Hornsgatan 1, 415 50 Gothenburg, Sweden, michael.wendel@skf.com (Corresponding author/Kontakt)
Prof. Dr.-Ing. habil. Franz Hoffmann, IWT Foundation Institute of Materials Science, Bremen, University Bremen, Germany
Walter Datchary, AB SKF, Gothenburg, Sweden

1 Introduction • Comparatively low specific energy consumption and environ-


mental impact (clean heating, lower CO2 emissions)
Besides heating for forming, joining, assembly or melting, induc- • Reproducibility
tion heating can be used for surface hardening, through hardening • Good integration in the production line (in line equipment)
or tempering of hardened steel components. Induction heating • On/ off equipment (low standby costs)
can provide advantages over other conventional heating technol- • High level of flexibility and automation.
ogies like [1]: For the bearing industry induction surface hardening, induction
• Heat generation directly within the workpiece through hardening, and induction tempering of components like
• Fast heating due to high energy density rolling elements or rings are established in production and are fur-
• High temperatures if required ther developed in research and development. Figure 1 shows main
• Demanded temperature distributions within the workpiece/ lo- factors impacting the result of an induction hardening process.
calized heating The steels chemistry, its prior structure and prior processing play

20 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening

Fig. 1.  Major factors impacting


Induction Hardening result

Bild 1.  Einflussfaktoren auf das


HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

Ergebnis der Induktiven Härtung


For personal use only.

important roles apart from the hardening process parameters and Beside the steel’s chemical composition, its prior structure before in-
the induction machine design. duction hardening plays an important role for the kinetics during
Induction heating or hardening processes are mostly char- heating and austenitisation and the subsequent quenching process. In
acterized by shorter processing times and higher temperatures general it can be stated that coarse microstructures with big grains and
compared to conventional heating methods. The limited time at long diffusion or carbide/cementite distances like soft annealed struc-
elevated temperature results in lower degrees of homogenization tures (SA) are unfavourable from an induction hardening point of view.
compared to furnace heating, especially for slower diffusing, sub- They require longer diffusion distances and lead in general to higher
stitutional dissolved alloying elements. The alloying itself as well transformation temperatures or longer times until transformation is
as the distribution of alloying elements within the microstructure completed and a homogenized austenite is formed [3, 8, 9]. High tem-
prior to induction heating play therefore an important role for the peratures and long times for austenitisation might lead to undesired
hardening result. hardening results including grain growth, coarse martensite, large tran-
The solubility and diffusivity of carbon or other alloying el- sition zones, surface damaging like decarburization and distortion [6].
ements are decreased by the presence of alloying elements like Structures with a small grain size and a fine distribution of carbides
chromium, hindering a fast austenitisation [2-5]. Therefore, the and cementite like quench and tempered (QT) or bainitised struc-
use of steels alloyed with high amounts of, for example, chromi- tures show advantageous hardening response. They assure a fast heat-
um, molybdenum, or vanadium is questionable from a heating ing response and allow a reduction of temperature or time resulting
response perspective. They diffuse slowly, dissolve preferably to a in less energy consumption, less grain growth, surface damaging and
large extend within stable carbides increasing their stability and distortion, possibly higher hardness and a deeper hardened case. Sur-
slow down the diffusion of, for example, carbon. However elements face hardened components with QT prior microstructures also show
like chromium or molybdenum increase hardenability and stable narrow transition zones, high core hardness but less core toughness as
carbides provide grain growth control. compared to soft annealed (SA) prior microstructures [6].
The steels chemistry, as well as its prior microstructure, affects However, it is not only the heat treatment deciding the proper
the phase transformation from ferrite to austenite during austen- starting structure. Earlier steps in the manufacturing chain like
itisation. The effect of alloying elements on the onset and comple- cold forming (e. g. rolling) or machining (e. g. turning/drilling)
tion of the transformation to austenite are described in literature request a microstructure which provides good formability and/or
[6, 7] (carbon content C < 0.6 wt.-% for Ac3): machinability such as a soft annealed structure in general does. A
balance between heating response and related properties on the
Ac1 (°C) = 723 − 10.7 Mn − 16.9 Ni + 29.1 Si + (1) one hand (case and core) and costs for the prior microstructure
16.9 Cr + 290 As + 6.38 W and its machining and/or forming in earlier production steps (i. e.
soft machining, rolling) on the other hand needs to be considered.
Ac3 (°C) = 910 − 203 √(C) − 15.2 Ni + 44.7 Si + (2) Chemistry, prior structure and prior processing also affect the
104 V + 31.5 Mo + 13.1 W steels magnetic and electrical properties which in turn impact the

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 21
M. Wendel et al.: Bearing steels for induction hardening

Steel Average chemical composition / wt.-%

DIN Material number C Si Mn Cr Ni Mo Cu

Through hardening bearing steels (ISO 683-17:2014-10)

100Cr6 1.3505 0.99 0.25 0.35 1.48 - max 0.1 max 0.3

100CrMnSi6-4 1.3520 0.99 0.60 1.10 1.53 - max 0.1 max 0.3

100CrMo7-3 1.3536 0.99 0.30 0.70 1.80 - 0.28 max 0.3

100CrMo7 1.3537 0.99 0.30 0.35 1.80 - 0.23 max 0.3

100CrMnMoSi8-4-6 1.3539 0.99 0.50 0.95 1.93 - 0.55 max 0.3


HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

Steels for quenching and tempering/plain carbon steels (DIN EN 10083-2:2006, DIN EN 10083-3:2006)
1.1151 ... 0.21 ... 0.55 ...
C22E … C60E max 0.4 max 0.4 max 0.4 max 0.1 -
1.1221 0.61 0.75
46Cr2 1.7006 0.46 max 0.4 0.65 0.50 - - -

41Cr4 1.7035 0.42 max 0.4 0.75 1.05 - - -


25CrMo4 … 1.7218 ... 0.26 ... 0.75 ...
max 0.4 1.05 - 0.23 -
50CrMo4 1.7228 0.50 0.65
51CrV4 1.8159 0.51 max 0.4 0.90 1.05 - - -
For personal use only.

Steels for induction hardening (HWG Inductoheat1)

Cf 70 1.1249 0.72 0.25 0.28 - - - -

40Mn4 1.1157 0.40 0,38 0.95 - - - -

46MnSi4 1.5121 0.46 0.80 1.05 - - - -

53MnSi4 1.5141 0.54 0.90 1.00 - - - -

58CrV4 1.8161 0.59 0.27 0.90 1.05 - - -

Induction-hardening bearing steels (ISO 683-17:2014-10)

C56E2 1.1219 0.56 max 0.4 0.75 - - - max 0.3

56Mn4 1.1233 0.56 max 0.4 1.05 - - - max 0.3

70Mn4 1.1244 0.70 max 0.4 0.95 - - - max 0.3

43CrMo4 1.3563 0.43 max 0.4 0.75 1.05 - 0.23 max 0.3

Table 1.  Bearing steels, plain carbon steels and steels for quenching and tempering for induction- and flamehardening according to DIN EN 10083,
ISO 683-17, HWG Inductoheat1and Dr. Sommer Werkstofftechnik database, limited to carbon content > 0.4 %; list not exhaustive; “-“ = not specified

Tabelle 1. Wälzlagerstähle, un- und niedriglegierte Kohlenstoffstähle und Vergütungsstähle für das Induktions- und Flammhärten nach DIN EN 10083, ISO
683-17, HWG Inductoheat1 sowie nach der Dr. Sommer Werkstofftechnik-Datenbank; beschränkt auf C-Gehalte > 0,4 Gew.-%; Liste nicht vollständig;
“-“ = nicht angegeben

1
HWG Inductoheat GmbH. http://www.hwg-inductoheat.de/fileadmin/pdf/wii.pdf, 03.03.2011.

22 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening

induction hardening result [6, 10]. Those aspects will not be given austenitisation parameters are compared. Low changes of the men-
attention here. tioned properties with changed austenitisation conditions, as well
Steel grades for induction surface hardening or flame hard- as higher local hardenability or lower amounts of non-martensitic
ening in general can be found in the standards DIN 17212DIN transformation products after quenching are considered as favourable
EN 10083 or for bearing steels in ISO 683-17 which include plain from a process stability point of view. A good process stability does
carbon steels like grade DIN C55 or C60 and alloyed steel grades. not only result in less variation of, for example, hardness or MS of an
Through hardening bearing steels shall be mentioned here as well induction hardened component. It also indicates smaller gradients of
since they are in some cases used for surface induction hardening those properties within a component, taking into account the typical
(Table 1). temperature gradient when heating using induction.
The listed steel grades often cannot entirely fulfil the require-
ments of a bearing steel or specific product requirements like suf- 2 Experimental
ficient hardness, toughness, fatigue strenght, hardenability, cleanli-
ness, formability, machinability or availability. The ability to be, for 2.1  Steels and prior structures
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

instance, carburized or nitrided can be relevant when combination


processes are considered. Table 2 shows a list of commercially available steel grades which
The objective of this study is to benchmark potential and exist- were chosen as potential steel grades for induction hardening. Sil-
ing bearing steels in terms of their induction hardening response. icon and manganese alloyed steel grades and steels with carbon
Based on a literature study on alloying element effects, the avail- concentrations larger than 0,5 wt.-% are preferred grades. The main
ability of steels as well as on the today’s most commonly used steel reasons are shown in Table 3.
grades for induction hardened bearing components, six potential The first part of this publication (part I) shows the results for
and existing bearing steels are selected for this study. steel grades 100Cr6, Ovako 832 and 50CrMo4. A following second
The robustness regarding heat treatment result of the selected publication (part II) will complement the results by presenting the
steels and their prior conditions prior to induction hardening shall be results for 67SiCr5, 70Mn4 and Ovako 677L.
pointed out. Here, the sensitivity of martensite start temperature (MS), As a reference 100Cr6 was included as the most frequently used
hardness after quenching, prior austenite grain size, and appearance bearing steel. It is known as a steel grade, which can be suitable for
of non-martensitic transformation products, depending on different induction hardening. Schlicht showed that induction heat treatment
For personal use only.

Steel Supplier C Si Mn P S Cr Ni Mo Cu Al MS*/ °C MS**/ °C


100Cr6 DEW 0.97 0.21 0.44 0.011 0.007 1.51 0.11 0.05 0.110 0.006 227 ± 5 127 ± 26
Ovako 832 Ovako 0.88 0.69 1.57 0.020 0.007 1.46 0.13 0.08 0.204 0.029 189 ± 5 116 ± 21 Part I
50CrMo4 Sidenor 0.51 0.24 0.70 0.008 0.006 1.05 0.08 0.17 0.227 0.016 262 ± 5 288 ± 17
67SiCr5, 70Mn4, Ovako 677L Part II
Table 2.  Steels for dilatometer study; alloying contents in weight-% from melt; MS temperatures (*for reference heat treatment with dilatometer of as
delivered prior structures TA 860 °C, ts 10 min, t8-5 10 s ** for full dissolution of alloying elements, formulas from [11, 12])

Tabelle 2.  Ausgewählte Stähle für die Dilatometerstudie; Gehalte in Gew.-%, Schmelzanalyse; MS Temperaturen (*Referenzwärmebehandlung im Dilatome-
ter der Stähle im Anlieferungszustand TA 860 °C, ts 10 min, t8-5 10 s **für eine vollständige Lösung der Legierungselemente, Formeln aus [11, 12])

→ Sufficient hardness
Carbon > 0.5 wt.-% → Fast austenite transformation kinetics for higher C contents [26, 27]
→ Residual carbides for grain size control and RCF properties

→ Cost effective alloying element, low probability of future scarcity [28, 29]
→ Dissolves in ferrite matrix providing “fast” hardenability and solid solution hardening
Silicon →D oes not have a large influence on lowering MS temperature (minor impact on
retained austenite level and minor impact on cracking probability during quenching)
→ Provides tempering resistance

→ Cost effective alloying element


→ Dissolves to a large extend in ferrite matrix, depending on prior heat treatment, providing
Manganese
“fast” hardenability (e. g. share of 66 % dissolved in ferrite for a soft annealed 100Cr6 [30])
→ Lowers Ac1 temperature, accelerates austenitisation [21, 26]

Table 3.  Alloying strategy

Tabelle 3.  Legierungsstrategie

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 23
M. Wendel et al.: Bearing steels for induction hardening

100Cr6
195 ± 9 HV1 SA* 295 ± 2 HV1 TT = 660 °C QT
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

Table 4.  Microstructures of prior


condition variants; Nital 1.5 %,
20 s; TT = tempering tempera-

Ovako 832
ture, * = as delivered condition

Tabelle 4.  Gefügebilder der Aus-


gangsgefüge; Ätzung in Nital
1.5 %, 20 s; TT = Härtetempera-
tur, * = Anlieferungszustand

223 ± 4 HV1 SA* 296 ± 2 HV1 TT = 685 °C QT


For personal use only.

50CrMo4

272 ± 8 HV1 QT1* 287 ± 7 HV1 TT = 660 °C QT2

of rings made of 100Cr6 can lead to comparable properties in terms ing the average of 7 different formulas compiled by Besserdich and
of hardness and strenghts compared to conventional isothermal Krauss [11, 12], see Table 3.
austenitisation, if soaking time was in the range of a few minutes [2]. The steels were investigated in different prior conditions accord-
Ovako 832 is a derivation of 100Cr6 and contents higher ing to Table 4. Besides the as delivered conditions, the steels were
amounts of silicon and manganese, expecting increased harden- heat treated to quenched and tough tempered states (QT). To set
ability. The lower carbon content could lead to a lower fraction the QT states, the steels were austenitised at 860 °C (30 minutes),
of carbides in the material and in total less alloying elements like quenched in 60 °C oil and subsequent tempered (2 hours) at var-
chromium and manganese bound to them. The share of alloying ious temperatures (see Table 4), aiming for a hardness of 300 HV.
elements dissolved in the matrix or in carbides however generally The tough tempered state is a compromise between response to
strongly depends on the prior treatment (e. g. soft annealing). Ova- induction hardening and machinability or formability. For grade
ko 832 is in use as a bearing steel grade. 50CrMo4 a second quench and tempered state QT2 was investigat-
50CrMo4 is used as a bearing steel and contains the lowest ed. The intention was to avoid the phases which appeared as lighter
amount of carbon. It contents a moderate amount of chromium etched areas on the micrograph of the QT1 structure (possibly tem-
and comparably high amount of molybdenum. Due to the low pered bainite fractions) and examine their effect on the heating and
amount of carbon and low amount of alloying elements lowering quenching response.
MS, the steel is assumed to be least critical regarding overheating
leading to high amounts of retained austenite after hardening 2.2  Dilatometer study
(highest theoretical MS for full dissolution).
Temperatures for the onset of martensitic transformation (MS) In order to understand the induction hardening response of the
for the case of full alloying element dissolution were calculated us- different steels and prior structures, a dilatometer study was

24 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening

heating rate/ K/s TA/ °C tS/ s 3 Quenching kinetics, as quenched


hardness, prior austenite grain size,
30 950 1 non-martensitic transformation
300 1000 10
products – results
Table 7, Table 8 and Table 9 show the as quenched hardness and
Table 5.  Heating parameters
measured MS temperatures of 100Cr6, Ovako 832 and 50CrMo4.
Tabelle 5.  Parameter der Erwärmung und Haltezeiten As an example, prior austenite grain sizes (PAGSs) and dilatation
curves in the temperature regime of martensite transformation
during quenching are shown for 100Cr6 in Table 8.
carried out. The induction heat treatment of hollow cylinders Table 10 shows micrographs of dilatometer samples in SA
(Øo 4 mm, length 10 mm, wall thickness 0.5 mm) was done with condition, heat treated with the lowest and highest austenitisa-
a dilatometer (type Bähr 805 A). The cylinders were heated with tion degree. For 100Cr6 NMTPs (non-martensitic transformation
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

a frequency of 2 MHz and a power of maximum 1.75 kW inside a products) are visible. The dilatation curves for the lowest austeniti-
coil (Øi 17 mm, length 24 mm, windings 6). The temperature was sation degree (300 K/s, 950 °C, 1 s) of 100Cr6 SA showed a deflec-
measured with a thermocouple (type S) spot welded on the outer tion during quenching at temperatures around 550–650 °C, typi-
cylinder surface. After heating with a certain heating rate to the cally indicating a diffusion controlled transformation of austenite
final temperature TA and soaking for a certain time tS, all cylinders (NMTPs). The amount of NMTPs on the micrographs of 100Cr6
were quenched with gaseous nitrogen (t8-5 = 5 s). The heating pa- SA was clearly decreased for higher austenitisation degrees, how-
rameters are shown in Table 5. ever NMTPs could not be fully avoided.
For the used dilatometer arrangement and sample geometry, Table 11 shows micrographs of dilatometer samples in QT
a programmed heating rate of 300 K/s could not be maintained condition, heat treated with the lowest and highest austenitisation
between the onset and completion of austenite transformation degree. 100Cr6 shows very little and small spots of NMTPs, which
(Ac1/Ac1b and Ac3/Ac1e) although using full generator power. Pos- disappear for higher austenitisations degrees. Ovako 832 QT and
sible reasons are the loss of magnetic properties from Curie tem- 50CrMo4 QT did not reveal any NMTPs. 50CrMo4 QT2 micro-
perature TC (lower efficiency), increase of currents penetration graphs were similar to the ones of 50CrMo4 QT1 and are therefore
For personal use only.

depth, heat losses due to crystallization and reaction speed of the not shown.
control loop. The measured heating rates between Ac1/Ac1b and
Ac3/Ac1e as well as the measured transformation temperatures 4 Quenching kinetics, as quenched
are given in Table 6. hardness, prior austenite grain size,
The dilatometer curves were evaluated regarding beginning non-martensitic transformation
of transformation from austenite to martensite (quenching) using products - discussion
the local minima of the dilatation curves. The heat treated steel
samples were characterized regarding hardness (8 indents HV1) In order to assess the robustness of the selected steels and their
and prior austenite grain size (geometric method, interception of prior microstructure regarding the induction heat treatment
grain boundaries with lines, 3 pictures evaluated), one sample per result, the arithmetic mean change in MS, as quenched hardness
parameter setting. The calculated average length of interception and prior austenite grain size for changed austenitisation condi-
line segments is referred to as prior austenite grain size in the fol- tions are shown in Figures 2, 3, and 4. Figure 5 shows the prior aus-
lowing. For the prior austenite grain size determination, the heat tenite grain size mapping, including all investigated steel grades
treated samples were etched in a mixture of picric acid, hydrochlo- (Bearing steels for induction hardening – parts I and II).
ride acid, and a wetting agent heated to approximately 60 °C.

Measured Measured
Ac1/Ac1b- Ac1/Ac1b- Ac1/Ac1b- Ac1/Ac1b-
Prior heating rate Prior heating rate
Steel Ac3/Ac1e Ac3/Ac1e Ac3/Ac1e Ac3/Ac1e
structure Ac1/Ac1b-Ac3/Ac1e structure Ac1/ Ac1b-Ac3/ Ac1e
/ °C ± 3 °C / °C ± 3 °C / °C ± 3 °C / °C ± 3 °C
(programmed) / K/s (programmed) / K/s
100Cr6 SA 30 (30), 133 (300) 795-846 825-859 QT 30 (30), 115 (300) 774-811 794-822

Ovako 832 SA 30 (30), 144 (300) 797-844 823-858 QT 30 (30), 141 (300) 779-825 798-837

50CrMo4 QT1 30 (30), 91 (300) 785-816 810-831 QT2 30 (30), 104 (300) 778-817 799-828

Table 6.  Programmed and measured heating rates and transformation temperatures during dilatometer heating; temperatures in grey indicate insufficient
heating rate

Tabelle 6. Programmierte und gemessene Erwärmgeschwindigkeiten im Temperaturbereich Ac1/Ac1b-Ac3/Ac1e, und Umwandlungstemperaturen während
der Erwärmung; Temperaturen in grauer Farbe zeigen zu geringe Erwärmgeschwindigkeit

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 25
M. Wendel et al.: Bearing steels for induction hardening
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

(1) Dilatometer curves for quenching from different austenitisation condi- (2) Dilatometer curves for quenching from different austenitisation condi-
tions - 100Cr6 SA tions - 100Cr6 QT
For personal use only.

(3) MS temperatures determined from dilatometer curves (local minima) (4) Hardness depending on austenitisation condition; maximum, carbon
for different austenitisation degrees - 100Cr6 depending martensite hardness according to [31] - 100Cr6

(5) Hardness – MS values and theoretical MS for full dissolution of all alloy- (6) Prior austenite grain size depending on prior structure and austenitisa-
ing elements [11, 12]; error bars show standard deviation - 100Cr6 tion conditions - 100Cr6

Table 7.  Quenching kinetics, as quenched hardness, prior austenite grain size – 100Cr6

Tabelle 7.  Kinetik während der Abschreckung, Härte und ehem. Austenitkorngröße – 100Cr6

26 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

(1) MS temperatures determined from dilatometer curves (local minima) (1) MS temperatures determined from dilatometer curves (local minima)
for different austenitisation degrees - Ovako 832 for different austenitisation degrees -50CrMo4
For personal use only.

(2) Hardness depending on austenitisation condition; maximum, carbon (2) Hardness depending on austenitisation condition; maximum, carbon
depending martensite hardness according to [31] - Ovako 832 depending martensite hardness according to [31] - 50CrMo4

(3) Hardness – MS values and theoretical MS for full dissolution of all allo- (3) Hardness – MS values and theoretical MS for full dissolution of all
ying elements [11, 12]; error bars show standard deviation - Ovako 832 alloying elements [11, 12]; error bars show standard deviation - 50CrMo4

Table 8.  Quenching kinetics, as quenched hardness, prior austenite grain Table 9.  Quenching kinetics, as quenched hardness, prior austenite grain
size – Ovako 832 size – 50CrMo4

Tabelle 8.  Kinetik während der Abschreckung, Härte- und ehem. Austenit- Tabelle 9.  Kinetik während der Abschreckung, Härte und ehem. Austenit-
korngröße – Ovako 832 korngröße – 50CrMo4

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 27
M. Wendel et al.: Bearing steels for induction hardening

100Cr6 SA Ovako 832 SA

heating rate 300 K/s


TA 950 °C Table 10.  . Microstructures for
t s 1s lowest and highest austenitisati-
on degree; etched in 1.5 % Nital,
25 seconds, SA/ N conditions;
scale = 10 µm

Tabelle 10.  Gefüge für geringste


und höchste Austenitisierung; Ät-
heating rate 30 K/s zung in Nital 1.5 %, 25 Sekunden;
TA
GKZ geglühte Ausgangsgefüge
1000 °C
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

t s 10 s

100Cr6 QT Ovako 832 QT 50CrMo4 QT1

heating rate 300 K/s


TA 950 °C
t s 1s
For personal use only.

heating rate 30 K/s


TA 1000 °C
t s 10 s

Table 11.  Microstructures for lowest and highest austenitisation degree; etched in 1.5 % Nital, 25 seconds, QT

Tabelle 11.  Gefüge für geringste und höchste Austenitisierung; Ätzung in Nital 1.5 %, 25 Sekunden; vorvergütete Ausgangsgefüge

4.1 100Cr6 wering alloying elements might be able to diffuse and dissolve more
elements into the austenitic matrix at the same time or temperatu-
4.1.1 MS re. On the other hand other features like carbides thermal stability,
austenitisation kinetics or other microstructural features like carbide
Ongoing dissolution of carbides and alloying elements into the aus- size, shape, spacing, density, individual phase chemistry or grain size
tenitic matrix leads to decreasing MS temperatures for increased might influence the dissolution kinetics as well.
austenitisation. Carbide size and carbide number on micrographs of Steels with fast changing MS temperatures, containing higher
the relevant dilatometer samples are decreasing with increased aus- amounts of MS lowering alloying elements, are more difficult to
tenitisation degree. The QT structures show lower MS temperatures control in terms of a desired MS in the given austentisation range.
due to faster austenitisation kinetics. Lower, measured transformation Too low MS temperatures bear the risk of excessive retained aus-
temperatures Ac1/Ac1b and Ac3/Ac1e for the QT prior structure, as well tenite contents after quenching which might cause severe problems
as microstructural features as carbide size, shape, spacing, density, in- in terms of surface finishing (e. g. grinding), dimensional stability
dividual phase chemistry or grain size are likely causes. during operation or hardness loss. Retained austenite content as
Generally, the magnitudes of TA and tS effects on MS (arithmetic well as thermal and mechanical stability of retained austenite are
mean change in MS when increasing or prolonging TA and tS) can important and might be very different for the investigated steel
be correlated to the steels theoretical MS for full dissolution of all al- grades. Contrarily, higher retained austenite contents might be de-
loying elements, see Table 3. The more negative the effects on MS, the sirable for the performance of certain bearing applications [13].
lower the calculated theoretical MS temperatures. This seems reas- The effect of tS on MS appears larger for most steels than of TA
onable on the one hand since steels with higher amounts of MS lo- on MS. The change in MS is connected to the dissolution of car-

28 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

Fig. 2.  Effects of heating rate, austenitisation temperature TA and soaking Fig. 3.  Prior austenite grain size mapping and steels molybdenum and alu-
time tS on prior austenite grain size minum content as well as AC3/AC1e temperatures for 30 K/s heating rate;
error bars show max. and min. measured grain size
Bild 2.  Effekte der Erwärmgeschwindigkeit, Austenitisiertemperatur (TA)
und Haltezeit (tS) auf die ehemalige Austenitkorngröße Bild 3.  Ehemalige Austenitkorngrößen, Molybdän- und Aluminiumgehalt,
sowie Ac3/Ac1e Temperaturen für eine Erwärmgeschwindigkeit von 30 K/s;
Fehlerbalken zeigen maximal und minimal gemessene ehem. Austenitkorn-
größe
For personal use only.

Fig. 5.  Isothermal carbide dissolution for different austenitisation tempera-


tures of a C85 steel from Grosch [17] (left) and derived dissolution for a
Fig. 4. Volume diffusion distances of chromium and carbon in austenite for
constant soaking time and increasing austenitisation temperature (right)
different times
Bild 5.  Isothermische Carbidauflösung für unterschiedliche Austenitisier-
Bild 4.  Diffusionswege Volumendiffusion für Chrom und Kohlenstoff im
temperaturen eines C85 nach Grosch [17] (links), sowie abgeleitet die
Austenit
Auflösung für konstante Haltezeiten und zunehmende Austenitisiertempe-
ratur (rechts)

bides and alloying elements into the austenitic matrix. Whether of alloying elements in the austenitic matrix. Beswick, Grosch and
the diffusion rate of carbon or other alloying elements is the con- Molinder show experimentally in [16-18] a decrease in carbide
trolling mechanism for carbide dissolution is discussed controver- dissolution rate for long soaking times at constant austenitisation
sial in literature, as summarized in [14]. An estimation for volume temperatures. The authors demonstrate that the decrease of car-
diffusion under simplified circumstances is given by Shewmon in bide fraction over soaking time of a bearing steel and low alloyed
[15], see Figure 6. It shows a greater effect of an increased soaking high carbon steels (100Cr6, C85, 1.27 % C steel) follows a sigmoid
time (1 s → 10 s) on the diffusion distance of carbon or chromi- function (”S function”), see Figure 7. Slowed down carbide disso-
um as compared to an increased temperature (950 °C → 1000 °C). lution kinetics for lower carbide fractions may have two possible
Hence from a diffusion distance point of view, a larger effect of reasons. Due to the faster austenitisation kinetics for the QT prior
tS could have been expected. 100Cr6 QT is an exception from the structure, higher amounts of alloying elements are dissolved in
mentioned observation. Other mechanisms apart from diffusion austenite for the same austenitisation condition compared to SA
rate or distance might be controlling mechanisms, like for example prior structures. The driving force for diffusion, the difference in
the thermal stability of alloyed carbides. elements potential between the alloying element sources (e. g. car-
The change in MS when changing TA and tS is smaller for the bides) and the alloying element sink (austenitic matrix), decreases
QT prior condition compared to the SA condition (smaller effect with ongoing element dissolution in austenite. Since the QT struc-
of TA and tS on MS). A decrease in MS for higher austenitisation tures` austenitic matrix contents a higher amount of alloying ele-
degrees is connected to dissolution of carbides and dissolution ments at the same austenitisation condition and hence exhibits a

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 29
M. Wendel et al.: Bearing steels for induction hardening

higher matrix potential and smaller potential difference, a further


increase in temperature or time leads to slower alloying element
uptake by the matrix.
Another approach takes different types of carbides into ac-
count. While Fe3C carbides dissolve comparably fast at the begin-
ning of the austenitisation process, more complex carbides alloyed
with chromium, manganese or molybdenum are more stable and
dissolve slower. A slowed down dissolution at higher temperatures
or longer times (higher austenitisation degrees) would be the re-
sult, hence a slower drop in MS. Since the austenitisation degree in
case of identical temperature and soaking time is higher for the
QT structure compared to the SA structure, a slower dissolution
and drop in MS temperature would be expected.
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

An increased heating rate leads to very small effects on MS in


Fig. 6. Volume diffusion distances of chromium and carbon in austenite for
case of 100Cr6 as well as for all steel grades. Earlier measurements different times
with a relatively high number of repetitions showed that MS can
differ in the range of ~ 5 °C for identical austenitisation condi- Bild 6.  Diffusionswege Volumendiffusion für Chrom und Kohlenstoff im
Austenit
tions. Since the effects are in the range of –5 °C up to +5 °C for
all steels, the significance is thought to be very low. Nevertheless
tendencies are tried to be elaborated. hardness. Both approaches will lower hardness whereas MS would
The negative effects of heating rate on MS of 100Cr6 might part- be shifted in negative (appearance of NMTPs) and positive (less
ly be caused by the increased appearance of NMTPs on the micro- time for dissolution) direction. The decrease of NMTPs and the
graphs for higher heating rates. The dilatation curves for the lowest ongoing dissolution of carbon for the case of 100Cr6 SA result in
austenitisation degree (950 °C, 1 s) of 100Cr6 SA (see Table 7 (1)), comparably high effects of TA and tS on the measured hardness.
shifted to higher dilatation, show a diffusion controlled transforma- NMTPs were observed to a much smaller extend in the case
tion at higher temperatures resulting in a certain amount of NMTPs. of 100Cr6 QT which might be one reason for a smaller negative
For personal use only.

The diffusion controlled transformation during quenching at higher effect on hardness for faster heating. The effects of TA and tS on
temperatures can lead to an enrichment of the adjacent austenite hardness are small. An increased austenitisation degree and con-
matrix with alloying elements, lowering MS. nected ongoing alloying element dissolution (see effects on MS)
lowers hardness from a MS temperature of ~ 195–200 °C, most
4.1.2 Hardness probably due to an increase in retained austenite content, see also
Table 7 (4), (5). Hence 100Cr6 is easily overheatable in terms of
The effects of heating rate on hardness of 100Cr6 SA and QT are retained austenite in the given austenitisation range, whereas
slightly negative. The negative effect of 100Cr6 SA appears larger 100Cr6 QT overheats at lower austenitisation degrees. The high-
and might partly be caused by an increased amount of NMTPs, est hardness among all steel grades was measured for 100Cr6 QT
as observed on the micrographs and as derived from the negative with 903 HV1 (theoretical martensite hardness 902 HV).
effect on MS, as well as from increased standard deviations for The appearance of NMTPs for low austenitisation degrees
the hardness of fast heated samples. The shorter time available on the one hand and the risk of overheating for higher austeni-
for dissolution of carbon and other alloying elements in case of tisation degrees on the other hand, narrow the optimal process
higher heating rates might also contribute to the lower measured window.

Fig. 7.  Isothermal carbide dissolution for different austenitisation temperatures of a C85 steel from Grosch [17] (left) and derived dissolution for a cons-
tant soaking time and increasing austenitisation temperature (right)

Bild 7.  Isothermische Carbidauflösung für unterschiedliche Austenitisiertemperaturen eines C85 nach Grosch [17] (links), sowie abgeleitet die Auflösung
für konstante Haltezeiten und zunehmende Austenitisiertemperatur (rechts)

30 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening

4.1.3 PAGS the theoretical martensite hardness of 902 HV. The hardness level
is comparably low. This is another indication for higher shares of
The increase in PAGSs is comparably high and the absolute PAGS retained austenite which lowers the overall hardness.
level the second highest for 100Cr6. The highest carbon content The high share of MS lowering alloying elements makes Ova-
would suggest a high carbide share and grain boundary pinning ko 832 easily overheatable in terms of hardness loss and higher
effects. At the same time the Ac3/Ac1e temperatures are compara- retained austenite levels in the given austenitisation range, espe-
bly high for 100Cr6 SA and might also suggest being beneficial cially for the QT prior structure. Whether it is possible to exceed
for smaller PAGSs. Arguments which might underline the rel- the maximal measured hardness of ~ 820 HV1 with changed aus-
atively high PAGSs are the low amounts of aluminium and the tenitisation conditions remains unclear, but it seems unlikely that
low amount of molybdenum which can reduce PAGSs as shown hardness levels of 100Cr6 can be reached.
by Hayashi et al. [19] and stated by Rudnev [6]. From the results Excessive retained austenite contents after hardening can be de-
obtained, 100Cr6 is thought to be less beneficial regarding PAGSs sirable for certain applications or unwanted with respect to phase
among the other steel grades. changes and dimensional changes during operation. The thermal
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

stability of retained austenite during tempering or tempering resis-


4.2  Ovako 832 tance in general influences the remaining amount of retained aus-
tenite for certain tempering conditions. Elements as silicon, chromi-
4.2.1 MS um, molybdenum or manganese are known to enhance tempering
resistance [20-23]. Comparing the chemistry of 100Cr6 and Ovako
Ongoing dissolution of carbides and alloying elements into the 832, one could expect a higher tempering resistance for Ovako 832
austenitic matrix leads to decreasing MS temperatures for in- due to increased silicon and manganese contents.
creased austenitisation. Carbide size and number on micrographs
of the relevant dilatometer samples are decreasing with increased 4.2.3 PAGS
austenitisation degree. The QT structures show lower MS tempera-
tures due to faster austenitisation kinetics. Lower transformation The average PAGS is smaller for Ovako 832 SA while the change
temperatures, as well as microstructural features as carbide size, in PAGS is larger compared to Ovako 832 QT. The reasons might
shape, spacing, density, individual phase chemistry or grain size be various and cannot be given in the frame of this work. Among
For personal use only.

are likely causes. others, the lower, measured Ac3/Ac1e temperatures for the QT prior
The change in MS when changing TA and tS is smaller for the structure might contribute to larger PAGSs.
QT pre condition compared to SA (smaller effect of TA and tS on The increasing effect of heating rate on PAGS of the QT version
MS). The same scenarios as for 100Cr6 seem reasonable for the de- is small and might be a result of the poor statistical coverage (only
scribed observation. one sample per factor setting) and the subjective evaluation of the
An increased heating rate leads to very small effects on MS. PAGS using the respective micrographs.
The negative or smaller effect on MS of Ovako 832 SA compared The average PAGSs of Ovako 832 appear to be smaller than
to Ovako 832 QT could be caused by a refinement of the prior aus- those of 70Mn4 and 100Cr6 and partly larger than those of
tenite grains. Ovako 832 SA shows the second largest, negative ef- 50CrMo4 and Ovako 677L. The same order is valid for the mo-
fect of heating rate on grain size (~ –2 µm), see Figure 4. However, lybdenum and aluminium content of the steels, whereas the alu-
since the effects are in the range of –5 °C up to +5 °C, the signifi- minium content of Ovako 832 is second largest after Ovako 677L
cance is thought to be very low. (Figure 5). This might be a hint that those elements partly play a
role and can affect the PAGS, as shown in [13, 19].
4.2.2 Hardness
4.3 50CrMo4
The effects on hardness for changed austenitisation degrees are small
in the case of Ovako 832 SA, although increased TA and prolonged 4.3.1 MS
tS let to the largest decrease in MS among all steel grades, see Fig-
ure 2. Looking at the hardness values in Table 8 (2), a small increase The transformation behaviour of 50CrMo4 shows little changes in
can be seen when increasing TA for the short soaked samples and MS for the different austenitisation degrees. A relatively small neg-
when prolonging tS at low TA. The reason for the small or missing ative effect of TA and tS on MS (see Figure 2) suggests that dissolu-
increase is most likely the increasing level of retained austenite after tion is ongoing but close to completion. Carbides are not visible on
hardening, which exhibits low hardness. Table 8 (2) and (3) show the micrographs (magnification 1000x) of the hardened dilatome-
that the overall hardness approaches relatively fast a stagnating or ter samples. The results of QT1 and QT2 are identical.
decreasing tendency for increasing austenitisation or decreasing MS. An increased heating rate leads to very small, negative effects on
As suspected from the lower absolute MS values of Ovako 832 MS. Assuming that no NMTPs are occurring during quenching and
QT, the tendency of losing hardness due to higher retained aus- full dissolution of carbides and alloying elements is not completed
tenite levels for increasing austenitisation is more pronounced. for the lowest austenitisation degree, an increase in MS would be ex-
This leads to negative effects for increased TA and prolonged tS. pected for faster heating. This is connected to a decreased time spent
The standard deviations of hardness measurements are increasing in the austenitic regime for higher heating rates, which is caused by
with TA and tS as well, indicating larger shares of phase fractions higher transformation temperatures and less time spent to reach TA.
with different hardness levels (martensite and retained austenite). Shorter times spent in the austenitic regime shorten the time for car-
The maximum measured hardness of 821 HV1 is lower than bide and alloying element dissolution and hence increases MS.

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 31
M. Wendel et al.: Bearing steels for induction hardening

Negative effects of heating rate on MS could be caused by an zirconium (0.18 wt.-%).The author suggests that grain growth is
austenite grain refinement. Different authors like Yang or Gar- hampered by the suppression of carbon diffusion by molybdenum
cia-Junceda show in [24, 25] that MS can be lowered by a decrease in solid solution. The described effect might be one reason for the
in prior austenite grain size. The drop in MS becomes steeper for stable and small PAGSs of 50CrMo4, which contents the second
smaller grain sizes. Garcia-Junceda shows a more or less linear highest amount of molybdenum (0.17 wt.-%).
drop in MS of ~ 30 °C between grain sizes of ~ 3–10 µm, Yang a Generally, the prior austenite grain size is influenced by many
drop of ~ 30 °C between ~ 3–20 µm. parameters. From the material site, micro alloying (e. g. Al, Nb, Ti)
In the case of 50CrMo4, no NMTPs were observed and the and the whole steels prior processing play major roles.
prior austenite grain size was not changing much for an increased
heating rate (see Figure 4) as well as for the case of hardness (see 5 Summary
Figure 3). Since the mentioned changes in MS, hardness and PAGS
are small, no significant conclusion can be drawn. Table 12 summarises the individual assessment of the investigated
steels and their different prior structures regarding control of MS,
4.3.2 Hardness
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

hardness and retained austenite (RA), possibility to achieve dif-


ferent RA levels with similar overall hardness, remaining carbides
The low change in hardness or effects of heating rate, TA and tS after hardening, NMTPs, PAGS, and achievable hardness level. The
on hardness correspond well with the low effects on MS. Decreas- benchmark is only valid inside the investigated austenitisation and
ing MS for increased TA and tS leads to an increased hardness. quenching conditions as well as for the used steel grades.
The increase in hardness and decrease in MS is likely caused by Control of MS:
ongoing dissolution of carbon and other elements into the aus- • Small effects on MS as for 50CrMo4 make a process control inside
tenitic matrix. The given fact that hardness is not increasing with the given austenitisation conditions easier.
higher TA for long tS and not increasing with longer tS for higher • The lower MS for full dissolution of all alloying elements, the
TA, as well as the fact that the theoretical martensite hardness is higher the magnitude of TA and tS effects on MS.
reached and no residual carbides are visible on the micrographs, • Steels containing higher amounts of MS lowering alloying ele-
indicates that dissolution is close to completion or completed. The ments are more difficult to control in terms of a desired MS and
low amount of visible NMTPs displays the steels’ comparably high hence retained austenite content in the given austentisation range.
For personal use only.

hardenability for the given austenitisation conditions. At the same • 50CrMo4 shows a complete dissolution or is close to complete
time, 50CrMo4 shows the highest MS for a hardness of ~ 800 HV1 dissolution already for the lowest degree of austenitisation.
among all steel grades. This makes the steel grade interesting, if low • Comparing Ovako 832 with 100Cr6, the alloying with manga-
retained austenite levels are required after hardening or tempering nese and silicon lead to a more sensitive MS towards increased
effects during quenching (auto-tempering) are of interest. Due to austenitisation.
the given carbon content, 50CrMo4 is limited to lower hardness • QT starting microstructures show generally faster austeniti-
levels compared to the other investigated steel grades. 50CrMo4 sation kinetics but can lead to overheating in terms of low MS
shows the lowest measured maximum hardness (799 HV1) as well and hardness drop due to increased retained austenite levels for
as the lowest theoretical martensite hardness (765 HV). the given austenitisation range, as for 100Cr6 and Ovako 832.
Control of hardness:
4.3.3 PAGS • 50CrMo4 QT shows comparably little effects on hardness with-
out hardness losses due to increased retained austenite contents
Grade 50CrMo4 shows only small sensitivity to austenite grain growth and therefore thought to show a better process stability regarding
when changing austenitisation conditions on a comparably small ab- hardness.
solute grain size level. The results are considered being beneficial from • A decrease in hardness was measured for 100Cr6 and Ovako
a process stability and mechanical property point of view. 832 from MS temperatures of approximately 200 °C and below
The lower carbon content and resulting low amount of re- and make the steel grades sensitive to overheating in terms of
maining, on the micrographs visible, carbides after austenitisation excessive retained austenite levels and hardness loss; the low MS
as well as the relatively low Ac3/ Ac1e temperatures (only 70Mn4 temperatures for full element dissolution (see Table 2) forebode
N/ QT, 100Cr6 QT lower) would not suggest 50CrMo4 to show an the described behaviour.
advantageous behaviour regarding PAGS among the investigat- • The alloying with manganese and silicon led in the comparison
ed steels. The aluminium alloying connected to a possible grain of Ovako 832 with 100Cr6 to a lower achievable hardness.
boundary pinning network of AlN precipitations is comparable • Although 100Cr6 QT and Ovako 832 SA revealed small chang-
to the other steel grades and somewhat lower compared to Ovako es in hardness, one can expect changes in the microstructure in
832 and Ovako 677L. Hayashi et al. show in [19] that molybdenum terms of increasing retained austenite contents.
additions have a large impact on refining prior austenite grains Possibility to tailor RA level:
up to approximately 0.5 wt.-% molybdenum. The study was done • 100Cr6 and Ovako 832 provide the possibility to adjust differ-
on 0.5 wt.-% carbon steels, with a maximum addition of 1.2 wt.- ent microstructures on the same, individual hardness levels (e. g.
% molybdenum, and austenitisation temperatures of 900 °C and low/ high retained austenite content); the change in hardness
1000 °C. The comparison of different alloying strategies showed, around MS ~ 200 °C appears small.
that the impact of molybdenum (0.4 wt.-%) on the grain refine- Remaining carbides:
ment was larger as compared to vanadium (0.11 wt.-%), niobium • Ovako 832 and 100Cr6 show, on micrographs visible, remaining car-
(0.18 wt.-%), chromium (0.39 wt.-%), titanium (0.096 wt.-%) or bides for the highest austenitisation degree of QT prior structures.

32 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)
M. Wendel et al.: Bearing steels for induction hardening

Remaining
Possibility Possible
Control of Control of Control of carbides
to tailor NMTPs5 PAGS6 hardness
MS1 hardness2 RA level3 (highest aust.
RA level4 level
degree)

100Cr6 SA – – 0 + + – 0 +

QT 0 + – + + 0 0 +

Ovako 832 SA – + – + + + 0 0

QT – + – + + + + 0

50CrMo4 QT1 + + + – – + + –


HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

QT2 + + + – – + + –

67SiCr5

70Mn4 Part II

Ovako 677L

1
according to effect chart: − for effect > 35 °C, + for effect ≤ 10 °C
2
according to effect chart: − for effect > 70 HV 1, + for effect < 50 HV 1
3
MS ≥ 200 °C as a border for acceptable amounts of RA
4
Possibility to adjust different RA levels with similar overall hardness
5
Share of NMTPs on micrographs of hardened samples, indicating local hardenability
6
With focus on absolute PAGS rather than change in PAGS with austenitisation
For personal use only.

Table 12.  Benchmark of steels and prior structures regarding MS, hardness, retained austenite (RA), remaining carbides after hardening, NMTPs, PAGS,
and achievable hardness level

Tabelle 12.  Bewertung der Stähle und Ausgangsgefüge bezüglich MS, Härte, Restaustenit, Carbide nach dem Härten, nicht martensitischen Umwandlungs-
bestandteile des Gefüges, ehem. Austenitkorngröße, sowie erreichbare Härte

NMTPs: 3. Smith, K. C.; Wise, J. P.; Krauss, G.: A microstructural analysis of rapid austenite
formation in induction hardened steels. Proc. 5th ASM Heat Treatment and Sur-
• Higher fractions of NMTPs were present in 100Cr6 SA, lower
face Engineering Conf., 7-9.06.00, Gothenburg, Sweden, 2000, ASM int., Materials
fractions for 100Cr6 QT; Ovako 832 and 50CrMo4 showed no or Park, OH, USA, on CD, 2000
very little fractions of NMTPs and are thought to be higher in 4. Garcia, C. I.; Deardo, A. J.: Formation of austenite in 1.5 pct Mn steels. Metall.
hardenability for the given austenitisation range. trans. A 12 (1981) 3, p. 521-530, DOI: 10.1007/BF02648551
• The alloying with manganese and silicon led in the comparison 5. Paxton, H. W.: The Formation of Austenite. In: Transformation and Hardenability
of Ovako 832 with 100Cr6 to a clear decrease in NMTPs or in- in Steels, Symposium, 27-28.02.67, Ann Arbor, Michigan, USA. Climax Molybde-
crease in local hardenability. num Company of Michigan, 1967
6. Rudnev, V.; Loveless, D.; Cook, R.; Black, M.: Handbook of Induction Heating. Mar-
• 100Cr6 revealed a narrow process window limited by the appear-
cel Dekker, New York, USA, 2005
ance of NMTPs on one hand and hardness drop and low MS on 7. Andrews, K. W.: Empirical formulae for the calculation of some transformation
the other hand for the given hardening conditions. temperatures. JISI, 1965
PAGS: 8. Clarke, K.: The effect of heating rate and microstructural scale on austenite forma-
• 50CrMo4 shows advantageous behaviour due to the lowest tion, austenite homogenization and as-quenched microstructure in three induc-
change of PAGSs on a low absolute PAGS level. tion hardenable steels. PhD Thesis, Colorado School of Mines, USA, 2008
9. Schlicht, H.: Beitrag zur Theorie des schnellen Erwärmens und schnellen Abküh-
• Ovako 832 SA and 50CrMo4 QT showed the smallest measured
lens von Stahl. HTM Haerterei-Techn. Mitt. 29 (1974) 3, p. 184-192
PAGSs. 10. Bozorth, R. M.: Ferromagnetism. IEEE Press, Piscataway, 1993
• The effect of increased temperature TA on the change in PAGSs 11. Besserdich, G.: Untersuchung zur Eigenspannungs- und Verzugsausbildung beim
appears to be larger than for a prolonged soaking time tS, vice Abschrecken von Zylindern aus den Stählen 42CrMo4 und Ck45 unter Berück-
versa for the change in MS and hardness for the chosen variations. sichtigung der Umwandlungsplastizität. PhD Thesis, University Karlsruhe, 1993
12. Krauss, G.: Steels: Processing, Structure and Performance. ASM Int., Materials
References Park, Ohio, USA, 2005
1. Baake, E.; Nacke; B.: Efficient heating by electromagnetic sources in metallurgical 13. Bhadeshia, H. K. D. H.: Steels for bearings. Progress Mat. Sci. 57 (2012) 2, p. 268-
processes: recent applications and development trends. Proc. Int. Symp. Heating by 435, DOI: 10.1016/j.pmatsci.2011.06.002
electromagnetic sources, 18-21.05.10, Padua, Italy, University Padua (ed.), p. XXIII- 14. Brooks, C. R.: Principles of the austenitization of steels. Elsevier, London, GB, 1992
XXX, 2010 15. Shewmon, P.: Diffusion in Solids. 2nd ed., Wiley, Hoboken, 1989
2. Schlicht, H.; Schreiber, E.; Fahry, H.: Induktives Härten eines übereutektoiden 16. Beswick, J. M.: The effect of chromium in high carbon bearing steel. Metall. Trans.
Wälzlagerstahls. HTM Haerterei-Techn. Mitt. 23 (1968) 3, p. 232-237 A 18 (1987) 11, p. 1897-1906, DOI: 10.1007/BF02647019

HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung) 33
M. Wendel et al.: Bearing steels for induction hardening

17. Grosch, J.; Kocjancic, B.; Reichelt, G.: Geschwindigkeitsbestimmende Vorgänge bei 26. Eckstein, H.-J.: Wärmebehandlung von Stahl. VEB Deutscher Verlag f. Grundst-
der Kurzzeitaustenitisierung. HTM Haerterei-Techn. Mitt. 39 (1984) 5, p.199-205 offindustrie, Leipzig, 1969
18. Molinder, G.: A quantitative study of the formation of austenite and the solution of 27. Schmidtmann, E.; Brandis, H.: Beitrag zur Austenitbildung in unlegierten und
cementite at different austenitizing temperatures for a 1.27% C steel. Acta Metall. niedrig legierten untereutektoidischen Stählen. Stahleisen, Duesseldorf, 1959
4 (1956) 6, p. 565-571, DOI: 10.1016/0001-6160(56)90157-2 28. Wouters, H.; Bol, D.: Material Scarcity – an M2i study. Material innovation insti-
19. Hayashi, T.; Kurosawa, N.; Yamada K.: Method of Prior Austenite Grain Refining tute (M2i), Delft, NL, 2009
Using Induction Hardening. JFE technical report No. 15 (2010), p. 4-9 29. Diederen, A. M.: Metal minerals scarcity: A call for managed austerity and the
20. Houdremont, E.: Handbuch der Sonderstahlkunde. 2nd volume, Stahleisen, Dues- elements of hope. The Hague for Strategic Studies, The Hague, NL, 2009
seldorf, 1956 30. Andersson-Drugge, I.-M.; Lund, T.: A ferritic-pearlitic steel with eutectoid compo-
21. Bain, E. C.: Functions of the alloying elements in steel. U.S. Steel Corporation, sition. Creative use of bearing steels. ASTM, Philadelphia, USA, 1993
Pittsburgh, USA, 1939 31. Liedtke, D.: About the interrelation between carbon content in steels and the mar-
22. Furuharal, T.; Miyamoto, G.; Maki, T.: Refinement of cementite in tempered lath tensite hardness. Materialwissenschaft und Werkstofftechnik 34 (2003), p. 86-92
martensite. Proc. 17th IFHTSE Congr., 27-30.10.08, Kobe, Japan, 2008, p. 423-429
23. Miyamoto, G.; Oh, J. C.; Hono, K.; Furuhara, T.; Maki, T.: Effect of partitioning of
Mn and Si on the. g.owth kincetics of cementite in tempered Fe-0.6 mass% C mar-
tensite. Acta Mat. 55 (2007) 15, p. 5027-5038, DOI: 10.1016/j.actamat.2007.05.023 Bibliography
HTM Journal of Heat Treatment and Materials downloaded from www.hanser-elibrary.com by 191.181.53.133 on January 16, 2018

24. Yang, H.-S.; Bhadeshia, H. K. D. H.: Austenite grain size and the martensite-start DOI:10.3139/105.110277
temperature. Scripta Mat. 60 (2009) 7, p. 493-495, DOI: 10.1016/j.scripta- HTM J. Heat Treatm. Mat.
mat.2008.11.043
25. García-Junceda, A.; Capdevila, C.; Caballero, F.G.; García de Andrés, C.: Depen-
71 (2016) 1; page 20-34
dence of martensite start temperature on fine austenite grain size. Scripta Mat. 58 © Carl Hanser Verlag GmbH & Co. KG
(2008) 2, p. 134-137, DOI: 10.1016/j.scriptamat.2007.09.017 ISSN 1867-2493
For personal use only.

Bücher/Books
DIN-Taschenbuch 478: Härte und Ver- Härteprüfverfahren im Überblick
schleißfestigkeit von Beschichtungen – Die Bestimmung der Härte und Verschleißfestigkeit ist eine
Prüfverfahren. 1. Aufl., 2015, 624 Seiten, der wichtigsten Voraussetzungen für das Beurteilen der Wi-
Beuth Verlag, 151 €. – ISBN 978-3-410- derstandsfähigkeit von Beschichtungen gegen mechanische
24407-3 Beanspruchung. Die DIN SPEC 91064:2015-02 „Härte und
Verschleißfestigkeit von Beschichtungen“ beschreibt alle
Das DIN-Taschenbuch stellt erstmals alle wichtigen Härteprüfverfahren. Da jedes Verfahren seine
genormten Prüfverfahren zum Beurteilen spezifische Anwendung hat, kann ein ungeeignetes Verfah-
der Härte und Verschleißfestigkeit von Beschichtungen bei ren zu einer falschen Aussage führen. Hier ist der Anhang
unterschiedlicher Beanspruchung bereit. Der Anwender A der SPEC sehr hilfreich, der eine Übersicht über die ge-
verfügt mit diesem DIN-Taschenbuch in kompakter Form bräuchlichsten Verfahren gibt und damit den Anwender bei
über alle notwendigen Informationen, um das für seinen der Auswahl des für den Einzelfall am besten geeigneten
Zweck am besten geeignete Prüfverfahren auszuwählen und Prüfverfahrens unterstützt. Da die in der DIN SPEC 91064
normgerecht durchzuführen. Der Band enthält sowohl den wiedergegebenen Durchführungen und Zahlenangaben nur
grundlegenden DIN-Fachbericht DIN SPEC 91064 selbst als einen groben Überblick geben können, wird für Details auf
auch alle Normen, auf die in diesem Dokument verwiesen die jeweils gültigen Normen verwiesen. Red.
wird: Insgesamt sind das 27 DIN-(EN)-(ISO)-Normen im
Originaltext, verkleinert auf das Format A5.

34 HTM J. Heat Treatm. Mat. 71 (2016) 1 (formerly HTM Z. Werkst. Wärmebeh. Fertigung)