Major Factors in the Design of Induction Heating Equipment for Surface Hardening

Junaid A. Siddiqui* and Ghulam Rasool Mughal** College of Engineering Pakistan Air Force-Karachi Institute of Economics and Technology, Korangi Creek Karachi-75190 (Pakistan)
Received on October 3, 2007; Accepted on November 28, 2007 Abstract In this paper, major factors in design of high frequency induction-heating equipment for surface hardening purpose are discussed. After a brief theoretical review of the basic concepts involved in surface hardening, the influence of ratio current penetration depth to the workpiece diameter (d/) on the efficiency and the power induced in the workpiece are discussed. Different types of hardening and quenching techniques, which are to be given due considerations while designing the equipment, are also described in detail. Mathematical computations have been intentionally avoided to ease matters for non-specialist readers. In conclusion, factors affecting the inductor design and the design requirements for radio frequency power source are outlined. 1. INTRODUCTION Induction surface hardening is widely applied in transport, machine tool and metal industries, engaged in heat treatment of machining elements. The advantages that led to the widespread application of induction surface hardening include: rapid heating, low scaling, less machining, fast cycle time, precise control of temperature, localized heating, no decarburization and no large scale grain. Induction surface hardening is a heat treatment process used to increase the durability of machine tool elements subjected to high stresses. By durability is meant the improvement in the resistance to wear and greater torsion strength. Many types of steel are surface hardened with heat to increase toughness and resistance to wear. High quality alloy steel can be replaced by cheaper carbon steel which has been surface hardened by induction method. The induction method of hardening offers the possibility of confining the heat to the outer layer subjected to stresses without affecting the hardness of the core. A tough original core with hardened surface layer offers considerable mechanical and dynamic advantage. 2. INDUCTION HARDENING CONCEPTS In induction surface hardening only the outer surface layer is heated to a hardening temperature (7500C for steel) and then cooled immediately by means of spray of water, air or oil after a certain metallurgical allowable time. Not all the materials are
* Author for correspondence. E-mail: <jasiddiqui@cyber.net.pk> ** E mail: <mughal@pafkiet.edu.pk>

capable of being hardened. The hardenable material must contain alloy or carbon content as it plays a vital role in the build-up of the desired hardness. For adequate hardness, the carbon content present in the material should not be less than 0.35%. Steel, alloy steel and castings are some examples of the material capable of being hardened. Fig: 1 shows the basic arrangement for induction surface hardening comprising of an inductor made of hollow copper tubing which surrounds the workpiece to be hardened. and a power source supplying the inductor with AC frequency power. The magnetic field established in the inductor induces a voltage in the workpiece, which drives a current on the surface of the workpiece and heats the steel temperature, Increase in the work piece is both due to Joules heat and under certain circumstances due to losses that occur when the magnetic field reverses (hysteresis). Since the magnetic field changes with the frequency of the applied voltage, the current distribution over the cross-section of the work piece is not uniform. Instead the current density is maximum and concentrated on the upper surface decreases exponentially according to: I = I0 e-x/ .....(1) from the upper surface to the interior of the workpiece [Fig: 2]. At high frequencies the effect of the current being concentrated at the upper surface of the workpiece (skin effect] is

Fig. 1. Basic arrangement for induction hardening.

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

72

Current density

pendent. In surface hardening of ferromagnetic materials and steel below the Curie point, r increases in direct proportion to the increase in temperature and it can lie in the range between 1 and 103-104. For non-magnetic materials and steel above the Curie point (7800C), r approaches unity, thereby resulting in rapid increase of . This rapid increase of beyond the Curie point is especially useful, since it minimizes the danger of upper surface getting overheated. Fig. 3 shows as a function of the operating frequencies for different material. This rapid increase is especially visible in case of iron between 15 840 0C [3].

Surface distance
Penetration depth

Fig. 2. Exponential curve showing distribution of current density as a function of distance from the upper surface of the workpiece [1]. more pronounced so that the center of the core is practically free of current. The thickness of the layer in which the current is attenuated to 37% of its initial value at the surface is termed as current penetration depth . The heat distribution over the surface of the workpiece is dependent on the depth of penetration of electrical current. For practical purposes, equation = 503 /......(2) holds good. Here, (mm) represents the current penetration depth, ( m), the specific resistance, the relative permeability (mm2) and frequency, (Hz) of the power source required for hardening. Equation (1) indicates that for high frequencies and low resistivity of the material selected for hardening, the current penetration depth will also be low. Consequently, the heat distribution and hardening depth, which is a function of current penetration, will also be shallower and confined to the upper surface (skin effect). This property has been made useable in induction surface hardening where low penetration depths are desired and are obtained at high frequencies and high power densities. The skin effect alone, however does not determine the temperature distribution over the workpiece cross-section. As a result of conduction and depending upon heating time, a part of the heat also flows to the interior of the workpiece. The heating time selected, is therefore, very short so that the amount of heat that travels to the interior of the workpiece as a result of conduction is minimum. Given the dimensions and the hardening temperature, the hardening depth will depened on the steel quality, frequency, power density, heating time and the geometry of the workpiece [1, 3]. Current penetration depth is an important parameter in the design of induction hardening equipment, because it finally determines the depth of the upper surface layer being heated. Since the magnitude of the frequency of the power source and the parameters of the workpiece material determine the current penetration depth, they have a considerable influence in the design of induction hardening equipment including the power source. The material parameters and are strongly temperature de-

Operating frequency

Al: Aluminium; Cu: Copper; Fe: Iron; Ni: Nickel; B: Brass Fig. 3. Current penetration depth as a function of operating frequencies. 3. CHOICE OF FREQUENCY In the design of induction heating equipment, it is important to know the frequency range within which the equipment must operate. The suitable operating frequency depends on the work piece dimension and the desired hardening depth. The usual frequency range lies between 500Hz - 10 kHz medium frequency and 100 kHz - 2MHz high frequency range. The maximum energy transfer in induction hardening between inductor and workpiece occurs at low current penetration depth (i.e high frequency). However, it is to be noted that the thermal efficiency deteriorates with decreasing . According to Kretzmann [3], the frequency selected for hardening must therefore, be such that the current penetration depth does not exceed 1/8 times the workpiece diameter. The substitution of this size of the workpiece in equation 1, gives an expression: Fmin = 16106 / d2 Hz.......eq. 3. This expression can be used to find out the minimum frequency value [3]. The minimum frequency value sets a limit while designing the power source for induction hardening equipment. This limit can be exceeded if reasonable efficiency is to be obtained but it cannot be lowered. The effect of ratio d/ is shown in the Table: 1. The plot of fmin [3] against the workpiece dimensions gives the minimum values of frequencies for different types of materials and is shown in Fig: 4. For a definite frequency, there is also a minimum workpiece dimen-

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

73

sion at which the efficiency is the most reasonable. Table1 gives approximate formulae for calculating the minimum frequency corresponding to minimum workpiece diameter [1, 2]. Table 1. Influence of the ratio d/ on the efficiency. Workpiece dia Current Penetration Depth (d/) 8 6 4 2 1 0.4 0.1

Table 2. Minimum diameter of the workpiece for definite frequency range. d in mm Geometry of the workpiece f(KHz) Rectangular 2500 d2

25 50 70

100 100 50

400 400 25

900 1.7

Efficiency(%) 95 (Energy conversion)

85

65

30

10

1 Circular

15000 7.8 d2

3.9

6.5

13

4. INDUCED POWER Induced power is another factor, which influences the design of induction hardening equipment. For various geometrical shapes to be hardened the induced power is calculated by: P = 1,987 x 10-9 x H2 x A x m x (KW) eq. (4) where m= f (d/) is a factor depending on the actual shape, and dimensions of workpiece and the value of the current penetration depth. The plot of m=f (d/) for different shapes in Fig: 5 show that m exhibits maximum at a particular value of d/. [1]. For larger ratio of workpiece diameters to the current penetration depth d/, the power absorbed approaches nearly 100% [Table 2]. Hence, in practice attempts are made to keep the ratio d/ equal to or greater than 4. This means a power absorption of 65%, with the decreasing ratio, however, the power absorbed reduces so much that it nearly equals the radiation losses [3]. 5. HARDENING DEPTH Hardening depth is always greater than penetration depth of current because even after the heated layer is cooled, the heat penetrates still further as a result of conduction. Hardening depth is dependent on frequency, heating time and power densities. The heating time is further dependent on quenching techniques and quenching medium. Besides, hardening depth depends on factors, which vary partially during the heating cycle. The exact determination of hardening depth is, therefore, for the same reason not possible. However, curves have been developed which give practical results. Fig 6(a), and 6 (b) show curves for both single shot (static) and scanning (progressive) method of hardening with the help of which hardening depth for different power densities, heating time and feed rate can be found. It is obvious

Frequency Hz

Diameter

Induced power

Fe: Iron; M: Bronze; Cu: Copper Fig. 4. Minimum frequency in relation to the work piece diameter for different materials [2].

d/ (Ratio) Fig. 5. Effect of the ratio d/ on the induced power for different geometrical shapes [3].

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

74

kw 1.0 cm2
0.9 0.8 0.7
Power density

6. HARDENING TECHNIQUES The optimum results for a definite hardening job depend on the hardening techniques. While designing the equipment for induction hardening, care must be taken to select the techniques, which incur minimum costs per hardened workpiece under given electrical, thermal and metallurgical conditions. Basically, two methods of surface hardening are in use, the single shot (static) and scanning (progressive) method of hardening Fig: 8a, 8b. All other methods are either a combination or a modification of either of these two methods. In single shot (static) hardening the workpiece is placed inside a suitable designed inductor, the power is switched on for a predetermined time and the workpiece is removed and then quenched. The quench-head is often located below the inductor and sprays water into the workpiece at the end of the heating cycle. Single shot method is usually applied to harden smaller cross -sections of oblong workpieces such as hardening of bearing surfaces, collars, undercuts and fillets. In scanning operation the workpiece inductor are moved relatively close to each other. This method is particularly suited to the hardening of the whole surfaces of long shafts and pipes. With the inductor usually being fixed, while the workpiece moves, the heat is continuous and so is the quenching. The scanning speed is 2-60m/s. The scanning speed must be such that the time required for the movement of the workpiece into the cooling ring lies within the metallurgical limit. In general, the length of the hardening zone in scanning operation is greater than the length of the inductor and the cooling ring. In static-hardening the induc-

0.6 0.5 0.4 0.3 0.2 100 0.1 0 0 2 4 6 8 10 12 200 14

Hardening depth (mm)

Fig. 6. (a)Variation of hardening depth with power density [17].

800C = 0.75 mm

f = 1MHz 0 0 0.5 1.0 1.5 Hardening depth (mm) 2.0

f = 500 KHz; S = 00.75 mm Fig. 6. (b) Variation of hardening depth with density [13]. from the fig: 7 that by changing the heating time and power densities, hardening depths of 0.25m and 3m can be obtained. In case (a) power density of 2.5KW/cm2 will be required for a heating time of 0.30 s. In case (b), 0.4KW/cm2 are needed for a heating time of 10s.[4,5]. The surface temperatures achieved in these cases are 850 and 1100 0C, respectively. A comparison of these figures shows that larger hardening depths can be obtained with smaller high frequency power. Small power densities result in longer heating time till the hardening temperature is reached. During this long heating time the heat penetrates deeper into the workpiece, which gives a greater hardening depth after it, is cooled. Smaller hardening depth therefore requires larger high frequency energy source and results in high investition costs. [6, 4, 5].
Power density

Hardening depth (mm)

Fig. 7. Variation of hardening depth with the surface power density at a frequency of 1 MHz [5]. T= Heating time.

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

75

are used as buffers. The machine cycle is mainly determined by the hardening sequence. Stalling of the material flow i.e a lack of components on the incoming side and an accumulation of workpieces on the outgoing side trigger the cycle of the devices, without influencing the hardening results and no action has to be taken by the operator. Fully automatic hardening machines are usually applied for a throughput of one type only for very similar type of the workpieces, whereas the semi-automatic machines are mostly used for a medium throughput of a large variety of component type [4, 10]. In practice, machines have been developed which can be used for both static and scanning mode of operation for surface hardening. Especially two basic designs of machines have been usually adopted, namely the vertical type design and the rotary bank type. In the vertical type machine, Fig: 9 the workpiece is brought into the inductor and clamped on both sides. Heating of the workpiece takes place with or without rotation in static or scanning mode of operation. This design serves the purpose of hardening the workpieces such as shafts and bolts etc. Another feature of this design is that zonewise heating of the workpiece as required in the hardening of camshafts is possible. Here, the workpiece is moved with scanning speed. Rotary type machines, Fig: 10, have been designed for hardening of faces and corners of parts such as axle shaft and wheel shafts. Two parts can be processed at a time with a production rate of 280 parts per hour or even more. The machine has a turn-table with as many as six stations: In loading, reloading, checking stations to determine if parts are properly loaded, heating and quenching. The quench-head is integrated into the inductor etc. Machines are used for static hardening but can be equipped with vertical type scanning machines [10, 7]. 8. QUENCHING TECHNIQUES Besides frequency, heating time and power, hardening depth is also dependent on quenching techniques, medium of quenching and speed of quenching. Quenching techniques consideration in the design of induction hardening equipment are, therefore, as important as hardening techniques. In general, quenching techniques can be direct or indirect. Usually, direct method of pressure spray quenching technique is chosen as it directly adapts to the induction hardening set-ups. The quenching mechanism is built into the inductor, Fig. 11, so as to heat and quench concurrently. By adjusting the distance between the quench ring and the coil, the hardening depth can be influenced. A quench ring is built around a multiturn coil. It sprays water through the turns or it can be placed in the line. Inline construction (workpiece moves into the coil to be quenched) is preferred. In a separate method, workpiece is quenched in a bath of oil or water. Usually, bath is agitated, the degree depending on the steel and shape of the workpiece. The quenching medium water, oil, air or polymer based liquid is chosen to give the required metallurgical properties.

Quenching equipment should be designed so as to produce the desired results. For instance coupling clearance must be considered to ensure quench effectiveness. Generally in single shot or static hardening, coupling is quite close (0.06 in). Clearance may be increased greatly, if it is necessary to harden parts of more than one diameter. Since stream velocity of the quenchant drops rapidly as the quenchant stream lengthens, quench equipment must include a separate heat exchanger for the control of quenching temperature. Pressure control must also be provided to ensure optimum heat removal over the entire surface. Cooling like the heating rate and austenitising temperature in the surface zone must be uniform and at a rate consistent with the type of steel or geometry of the workpiece [11]. 9. INDUCTORS The inductor is the heart of induction hardening equipment producing a magnetic field, which is induced into the workpiece to be hardened. In surface hardening, the workpiece is very rapidly brought to a high temperature by means of high frequency currents, with current densities as much as 6000A/mm2. Such current densities therefore require that the material selected for inductors must possess high thermal conductance and low resistivity. Nearly all the inductors are made of hollow, water-cooled copper tubes of sufficient cross-section so as to give requisite mechanical strength and also to carry currents without getting overheated. The greater the current and power requirements, the greater the losses. The impedance of the inductor and with that the resistive losses are affected by various other factors such as turns, coil diameter and frequency. This means that a coil with a small diameter and low resistivity material (e.g. copper) and

Fig. 11. Inductors with built-in quenching mechanism.

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

77

operating at low frequency will result in minimum losses. However, greater power inputs result from higher frequencies, so the designer has to compromise between the most efficient type, the shape and the power requirement of the material to be hardened. In general high coil efficiencies (taking into account the power loading into the metal) are achieved at higher frequencies and large diameters. Coil design, therefore, must be such as to give the best heat pattern and highest degree of effciency. While designing inductors for surface hardening, the impedance of the power source must also be kept in mind. This is particularly important if the maximum available power from the energy source is to be utilized. Inductors are either conected directly to the resonant circuit of the energy source or through a radio frequency transformer (Fig. 12). The impedance appearing at the terminals of the single turn and multiturn inductor lies between 5-200 m and has to be matched to the impedance of the energy source. The form and shape of the workpiece to be hardened alongwith the contours of the hardening zones determine the design of the inductor. The coil or inductor must be designed to adapt exactly to the form of the workpiece and the path of the hardening zone so as to avoid undesireable heating zones and the heat flow resulting due to mutual effects of the magnetic fields, in case the inductor only approximately fits into the workpiece. Basically two inductors are practically in use: the internal and the external field inductor (Fig: 13), depending on whether the inductor is surrounded by the workpiece or the workpiece encircles the inductor. Complication arise when work pieces such as teethed gear are to be hardened in such cases both internal and external field configuration are incorporated in the design workpiece. Coupling is another factor, which must be taken care of while designing inductors. The current density distribution is dependent on the coupling i.e. the distance between the inductor, and the workpiece. Since magnetic fields are stronger near to the coil than at any distance away from it, it is advantageous to place the workpiece close to the inductor, so that the maximum of the heat energy is transferred to it. The strength of the field varies inversely with the square of the distance between the workpiece and the coil, which means that this consideration will have direct relation to the amount of heat generated in a workpiece in a given length of time. With multiturn coils closely coupled to the workpiece, there is a tendency for the eddy currents to provide a heat pattern corresponding to the helix of the coil. The wider the pitch of the coil, the more pronounced will be this heat pattern. Therefore, with a closely wound coil the rotation of the workpiece becomes essential. When the inductor is more loosely coupled i. e., at a greater distance from the surface to be heated, Fig: 14, the stream of eddy currents spreads over a wider area and rotation of the workpiece may not be necessary. . In surface hardening, the inductor currents are of higher frequencies and therefore, the current densities in the inductors will not be uniform due to skin effect. A further factor, which causes

departure from the uniformity, is the proximity effect, which will cause the bulk of the inductor currents to flow in that part of the conductor nearest to the workpiece. Although air insulation between adjacent turns of the coil is adequate from the electrical point of view, it does not always give the requisite mechanical rigidity for a coil used in production situation and spacers of suitable insulation material, which must be capable of withstanding the high temperatures, are used [7, 8, 3]. The production of coils for heating simple shapes to a uniform depth is straightforward, but designs are very complex in many instances such as heating of gears where the correct distribution of magnetic field is important. In many cases, it may be neces-

Fig. 12. HF - matching transformer.

(a) Internal

(b) External

Fig. 13 (a,b). Field distribution pattern of internal and external field inductors.

Fig. 13(c). Some designs of internal and external field inductors.

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

78

The DC terminals of the inverter are tightly coupled to RF bypass capacitor whose capacitance is sufficient to pass the AC component of inverter input without substantially changing its DC potential. The Ac terminals of the inverter drive the RF load circuit which is essentially a high Q series resonant circuit formed by tuning capacitor, C and inductive coil Lf. A radio frequency transformer matches the load impedance to the VA capability of the inverter, while coupling capacitor prevents any DC current from flowing in the primary winding and saturating the core. In operation, the MOSFET transistors are switched as diagonal pairs, Q1 and Q2 alternating each half cycle with Q3 and Q4 to provide a square wave voltage output at the AC terminals of the inverter. The waveform of the output current depends on the inverter frequency, which is the switching rate of the MOSFET transistors. Driving the series resonant load off resonance i.e. at a MOSFET switching frequency differing from the natural resonant frequency of the C1 and Lf results in a low output current, while driving it at resonance results in a maximum power to the load coil. Infact the output current is controlled in a closed loop by varying the driving frequency [14]. 10. REFERENCES [1] RWE-Essen: Inductive Erwaermung, Physikalische Grurundlagen und technische anwendungen, 2.Band 1979 Energie-Verlag, GmbH, and Heidelberg. W.Barth: Die anwendbarkeit und die Grenzen der induktives Haertung Teil:I Fertigungstechnik 7.Jg, Heft 6,june 1957 Benkowsky Induktives Erwaermung 4.bearbeitete Auflage, VEB-Verlag Berlin. , 1980 G.W.Seulen: An up-to-date look on induction hardening equipment.and F.H.Reinke Elektrowaerme International 31 (1973) B4 AEG-Elotherm: Induktives Randsicht heartens von Stahteilen Merkblatt 236 Kurt Flicke: Hochfrequenz-Induktionshaerteanlage DrahtWelt Ausgabe 1969.Nr.3, Seite 65-169. W.E. Mulane: Coil design for HF induction heating Metal treating Aug-sept.1963. P.G.Simpson: Induction heating, coil and system design. McGraw-Hill. F.W. Curtis High frequency Induction Heating McGrawHill, Newyork Oliver S. Park Single shot hardening by induction heating George Lendl: Why quenching is important in induction Heating? Metal progress Dec 1967 S.N.Okele Application of Thyristor inverters in induction heating and melting. Electronics and power March 1978. S.R.Pelly: Latest developments in static high frequency power sources for induction Heating. Solid -state Generator for induction hardening print from Conference record industry applications Society IEEE, IAS, annual meeting, October 4-7, 1982 , San Fransisco. G.W.Seulen: Entwicklungsstand der Induktionhaerte-

technik fuer Kurbelwelle Klepzig Fachberichte 1/72 [16] J. A. Siddiqui: Leistungselektronische Speisegeraete fuer die inductive erwaermung Unter besondere Beruecksichtigung der verschiedenen technologischen Verfahren unter der Netzanschluessprobleme. [Unpublished Thesis] [17] VDIArbeitsblatt Induktives Haerten Professor Dr. Ghulam Rasool Mughal CMILT Professor of electronics at the College of Engineering PAF-KIET, Korangi Creek Karachi received his B.Sc. (Hons.) in 1964, M. Sc (for Hons.) in 1965 from the University of Sindh, Jamshoro Sindh and PhD in 1974 from the University of Southampton, England. Prior joining the PAF-KIET, Dr. Mughal was professor of Microelectronics and an associate Dean at the Institute of Business and Technology, BIZTEK, Main Ibrahim Hydri Road, Karachi. Before adopting full time teaching profession at PAF-KIET and BIZTEK, Dr. Mughal was Chief Scientific Officer (BPS20), and served as a Head of the Research Division, Applied Physics, Computers and Instrumentation Technology Research Division, PCSIR Karachi Laboratories Complex, Karachi-75280. Dr. Mughal joined PCSIR Karachi Laboratories in June 1966 as Research Assistant and successively rose to the post of Chief Scientific Officer. The main task was full time scientific and Industrial Research. During his stay at PCSIR, he was visiting professor parttime, under the UGC (Higher Education Commission) teaching programme in the Department of Physics, University of Sindh, Jamshoro (1975-77) and taught of Microelectronics to M. Sc (Final) students. Dr. Mughal also went to Iraq and served as a visiting professor in the College of engineering and the College of Sciences, University of Basrah, Basrah, Iraq and taught Microelectronics and Physics to B.E. students (1980-82). Dr. Mughal assisted in establishing, Institute of Industrial Electronics Engineering (IIEE-PCSIR) faculty of NED University of Engineering & Technology. He taught Solid-state Devices Technology/Integrated Circuits & Physics to B. E students (1989-94). Simultaneously Dr. Mughal was teaching Industrial Electronics and Physics to the Post-diploma students at PSTC-PCSIR (1989-94). Dr. Mughal has 36 years R&D experience and 15 years teaching experience in Pakistan and abroad. He has also industrial experience in respect of repair and calibration of electronic gadgets. Dr. Mughal has 30 publications of international repute to his credit. He is member of a number of societies in Pakistan. Research Concentration and Areas of Subject Microelectronics, Physics & Technology of Solid-state Devices, Instrumentation and Education Delivery Systems in Science and Engineering Faculty.

[2]

[3] [4]

[5] [6] [7] [8] [9] [10] [11] [12] [13] [14]

[15]

TECHNOLOGY FORCES (Technol. forces): PAF-KIET Journal of Engineering and Sciences Volume 01, Number 02, July-December 2007

81