Numerical simulation as a tool for prediction of material behavior during induction quenching.
Behulova, Maria ; Taraba, Bohumil
Abstract: Induction heating followed by water quenching is widely
used for hardening of complex-shaped parts when hard surface and tough
part core are required. In this paper, the process of induction
hardening of a flanged plug from the steel C45E is analysed using
numerical simulation of coupled electro-magnetic, temperature and stress
fields. On the base of obtained results, the critical areas of possible
material failure are predicted.
Key words: induction heating, water spray cooling, numerical
simulation, temperature fields, stress-strain fields, crack.
1. INTRODUCTION
Induction quenching is widely used in industry. The main advantages
of this method include very accurate control of heated surface and
depth, fast heating rates, high efficiency, good reproducibility,
cleanness and low energy consumption.
Processes of induction quenching are mostly designed using
engineering experience and a trial-and-error procedure. In connection
with advance in computational technique and numerical methods, computer
simulation is ever more exploited to support design and optimisation of
induction heat treatment processes (H'omberg, 2004, Magnabosco, I.
et al., 2006).
2. PROBLEM DESCRIPTION
A flanged plug (Fig. 1) made from the steel C45E is under operation
subjected to the combined loading. In order to increase the hardness of
surface layers along with keeping relatively high core toughness, the
flanged plug is quenched using induction heating followed by water spray
cooling. During quenching, the plug is supported in centres and it is
rotating around its axis to achieve uniform through-hardening. The
hardened depth from 2 mm to 3.5 mm is required in the upper and middle
parts of a shaft. In the bottom conical part, the chill depth should be
from 2 mm to 5 mm, while the minimal hardened layer of 1.5 mm is desired
at the plug bottom. Designed shape of inductor coil was adapted to the
shape of plug bottom. Inductor is created by a copper single-turn coil
with dimension according to Fig. 1.
[FIGURE 1 OMITTED]
Technological process of induction quenching consists of several
steps. First, the inductor coil in off-state is fed down along the shaft
to the plug bottom. In this position, pre-heating of the area of plug
bottom is performed with the generator output of 23.8 kW and the
frequency of 20 kHz for a period of 6.9 seconds. Then, generator is
switch-off for the dwell time of 2 seconds. In the next step, inductor
coil is moving up along the shaft in the vertical direction with the
speed of 13.3 mm.[s.sup.-1] and generator output of 33.8 kW during 5.1
seconds. In the time of 9.9 sec., the process of cooling starts with the
water mass flow of 50 kg.[min.sup.-1] and temperature of 24 [degrees]C.
The paper deals with numerical simulation of the process of
induction quenching of a flanged plug using the finite element program
code ANSYS 10.0 in order to analyse electromagnetic, temperature and
stress-strain fields in a quenched part. The main aim of the paper is to
predict material behavior during induction quenching including the
analysis of the possibilities of heat treatment cracks and material
failure.
3. PRINCIPLE OF INDUCTION HEATING
In induction heating process, an alternating current is passed
through a conductor creating an alternating magnetic field. In a part
placed into this magnetic field, eddy currents are induced and material
is heated up due to the resistive and hysteresis looses. The induced
current and consequently the generated heat are not uniformly
distributed throughout the workpiece. Around 63 % of induced current and
87 % of generated heat is concentrated in the outer region called skin
depth depending on electric and magnetic properties and mainly on the
frequency of current in the coil (Haimbaugh, 2001). In the processes of
surface quenching of steels, the surface layer is heated very fast using
the high-frequency inductor whereby the internal parts are practically
unaffected. Rapid oil or water quenching results in the development of
thin hardened surface layers.
4. SIMULATION MODEL
The material behavior during induction heating and quenching is
very complex. It can be described by coupled sets of partial
differential equations, complicated boundary conditions for
electro-magnetic, thermal and stress-strain fields and nonlinear
material properties (Behulova & Taraba, 2006). The coupling between
electro-magnetic and temperature fields is given by the dependence of
electro-magnetic properties on temperature. On the other hand, the
temperature field is influenced by the Joule heat generation. In
addition, the magnetic permeability depends on the magnetic field
density and thermal properties are temperature dependent. The boundary
conditions, particularly the heat transfer coefficient varies very
strongly with the temperature. The temperature fields are further
influenced by movement of inductor and cooler. Moreover, a heated part
is deformed during induction heating due to the thermal expansion and
plastic deformation occurrence, i. e. the distance between inductor and
heated part is changed. The dependence of mechanical properties on
temperature must be taken into account as well. All these facts were
considered in simulation model and solution procedure using the finite
element program code ANSYS 10.0.
Based on the technical documentation of a flanged plug, the
axisymmetric geometrical model was prepared. The density of generated FE
mesh is considerably higher along quenched surfaces (Fig. 2) in order to
take into account the skin effect. Initial position of inductor is
illustrated in detail in Fig. 2b.
[FIGURE 2 OMITTED]
Material properties of the C45E steel for electro-magnetic, thermal
and static analyses were considered as functions of temperature.
Temperature dependent B-H curves were given by special user subroutine.
Initial and boundary conditions were defined in compliance with
described technological process of induction quenching. The initial
temperature of a plug was supposed 20 [degrees]C. The heat extraction
from the plug surface to the surroundings with the temperature of 20
[degrees]C is realised by free convection and radiation. The combined
heat transfer coefficient was calculated in the dependence on plug
surface temperature (Incropera & DeWitt, 1996). During water spray
cooling with temperature of 24 [degrees]C, the value of the heat
transfer coefficient of 5000 W.[m.sup.-2].[K.sup.-1] for was supposed
(Specht, 1992).
5. RESULTS AND DISSCUSION
Results of numerical simulation were analysed and evaluated with
the aim to predict the material behavior during induction quenching and
to examine critical regions of possible plug failure. In the phase of
pre-heating of a plug bottom, the maximum heating rates are reached in
the areas with the highest intensity of magnetic field. The maximum
temperatures at the end of pre-heating in the time of 6.9 sec. exceed
1100 [degrees]C (Fig. 3a). The maximum stresses at this time are
localised at the outer side of a plug under structural element called a
"cup" (Fig. 3b). The maximum displacements along the cup
circumference reach the values up to 0.34 mm. In the node 1, the
equivalent elastic Mises stresses ([[sigma].sub.M] = 430 MPa) exceed the
yield strength of the C45E steel ([R.sub.e,min] = 390 MPa) indicating
possibility of the occurrence of plastic deformations in the area of the
plug root.
During the holding time, temperatures in the critical parts of a
cup slightly decrease (Fig. 3d) mainly due to the heat conduction to the
inner parts of a plug and also the heat extraction to the surroundings.
In the time of 9.9 sec., the plug surface started to be water spray
quenched. Rapid temperature drop in the area of a cup results in the
enhacement of equivalent Mises stresses (Fig. 3d). Tensile stresses rise
in the superficial plug areas while compression stresses act inside the
plug. The maximum values of equivalent Misses stresses are moving from
cimcumference parts of the plug to the inner cup area attaining values
more than 1000 MPa (Fig. 3c). As these values exceed the ultimate
strength of the C45E steel ([R.sub.m] = 785 MPa), appearance of cracks
and material failure can be awaited with a high probability in the cup
area.
[FIGURE 3 OMITTED]
6. CONCLUSIONS
From the construction point of view, the analysed flanged plug
represents a complicated part comprising of different cross-sections
with variously specified hardened depths. Based on numerical simulation,
the area of a cup seems to be the most critical place of the possible
plug failure during induction quenching or later on operating
conditions. In this region, equivalent Misses stresses exceed the
ultimate strength of the C45E steel during quenching. Moreover, the cup
is hardened through the whole thickness what was proved experimentally
as well. This fact increase brittle damage ability of the cup.
The higher level of structural analysis represents the application
of elastic-plastic material model which enables to evaluate plastic
deformations and residual stresses. However, according to the
author's experience, the elastic model provides sometimes better
information on critical regions with potential occurrance of maximum
stresses and possible initiation of material damage during and/or after
heat treatment.
Nevertheless, the numerical simulation is a powerful tool for
examination of material behavior as well as for analysing and
optimisation of technological processes. In this case, the next work
will be focused particularly on the optimisation of inductor shape and
the process of induction heating.
The research has been supported by VEGA MS SR and SAV within the
projects No. 1/2073/05 and VEGA 1/2101/05.
7. REFERENCES
H'omberg, D. (2004). A mathematical model for induction
hardening including mechanical effects. Nonlinear Analysis: Real World
Applications 5, 2004, pp. 55-90.
Magnabosco, I. et al. (2006). Induction heat treatment of a ISO C45
steel bar: Experimental and numerical analysis. Computational Materials
Science 35, 2006, pp. 98-106.
Haimbaugh, R. E. (2001). Practical Induction Heat Treating, ASM International, Materials Park, Ohio.
Behulova, M. & Taraba, B. (2006) Analysis of temperature and
stress fields by induction hardening. In: 21st Int. Conf. on Heat
Treatment. ATZK, Jihlava, pp. 305-312.
Incropera, F. P. & DeWitt, D. P. (1996). Fundamentals of heat
and mass transfer, John Wiley & Sons, New York.
Specht, E. & Jeschar, R. (1992). Heat transfer in continuous
casting during spray-water cooling. Heat and Mass Transfer in Materials
Processing, pp. 535-547.