Figure 4: Tensile strength and hardness results of lamellar graphite cast iron materials produced at different casting temperatures

When the tensile strength and hardness results of lamellar graphite cast iron materials are examined, it has been observed that
with increasing casting temperature, the tensile strength and hardness values of the samples generally increase. However, the
strength properties of samples with a casting temperature of 1400oC showed a slight decrease.

Figure 5. Change in nodule quantity and size of spheroidal graphite cast iron samples depending on casting temperature

Figure 6. Change in phase amounts of casting temperature in cast iron depending on with spheroidal graphite

314
In spheroidal graphite cast irons, as shown in Figures 5 and 6, the amount of graphite nodule decreases with increasing
casting temperature. However, graphite nodule size has been observed to increase with increasing casting temperature. With
the increase in the casting temperature in the spheroidal graphite cast irons, an increase in the amount of graphite phase was
observed, but the amount of ferrite phase decreased. The amounts of perlite phases in the samples generally did not change.

Figure 7: Microstructure drawings of spheroidal graphite cast iron specimens produced at casting temperature 1325 oC a) GJS-A (x200), b)
GJS-B (x100), c) GJS-C(x200)

Figure 8: Tensile strength and hardness results of spherical graphite cast iron materials produced at different casting temperatures

When the tensile strength and hardness results of spherical graphite cast iron materials are examined, it has been observed
that increase in tensile strength and hardness values of the samples are generally observed when the increase in the casting
temperature. However, the casting temperature of the GJS-B coded sample showed a decrease in hardness value after
1350°C.

4. CONCLUSION
The results obtained from the materials produced by adjusting the casting temperature between 1300oC and 1400oC in
lamellas and spheroidal graphite cast irons with different chemical compositions are summarized below.

When the microstructure and mechanical properties of the materials produced at different casting temperatures are examined,
it has been understood that setting the casting temperature of the lamellar graphite cast iron materials at 1325oC to 1350oC is
suitable for the materials to be produced in the specified composition. However, it has been found that for the graphite cast
iron material with the chemical composition code GJS-A, the casting temperature is higher than 1400 ° C and for GJS-B and
GJS-C coded samples it is set to 1350°C to 1375°C.

For lamellar graphite cast iron, the general microstructure was found to be of type A 1 according to ASTM A247, but with
increasing casting temperature and an increase in the size of GLJ-B coded observed.

315
In the case of the increase of the casting temperature spheroidal graphite cast irons, the decrease of the number of nodules
have occurred. As a result, the amount of ferrite phase in the microstructure decreased and the amount of perlite and graphite
phase increase observed.

5. REFERENCES
[1]. Z. L. Lu, Y. X. Zhou, Q. C. Rao, Z. H. Jin, J. Mater. Process. Technol. 116 (2001) 176-181.
[2]. L. Ribeiro, A. Barbosa, F. Viana, A. Monteiro, Baptista, C. Dias, C. A. Ribeiro, Wear 270 (2011) 535-540.
[3]. Z. R. Yang, D. S. Li, S. Q. Wang, M. X. Wei, J. Iron Steel Res. Int. 20 (2013) No.10, 81-86.
[4]. Kumar S, Kumar P, Shan HS. Effect of Evaporative Pattern Casting Process Parameters on the Surface Roughness of Al-7% Si Alloy
Casting. Journal of Materials Processing Technology. 2007; 182: 615-623.
[5]. Kumar S, Kumar P, Shan HS. Optimation of Tensile Properties of Evaporative Casting Process Through Taguchi's Method. Journal of
Materials Processing Technology. 2008; 204:59-69.
[6]. Yao X, Shivkumar S. Molding filling characteristics in lost foam casting process. Materials science and Technology. 1997; 31: 841-
846.
[7]. Khodai M, Parvin N. Pressure Measurement and Some Lost Foam Casting in Observation. Journal of Materials Processing and
Technology. 2008; 206:1-8.
[8]. Shivkumar S, Yao X, Makhlouf M. Polymer Melt Interactions During Formation in the Lost Foam Process. Metallurgica et Materialia
scripta.1995; 33:39-46.
[9]. Shin SR, Lee ZH. Hydrogen Gas Pic-Up of Alloy During Lost Foam Casting Melt. Journal of Materials Science. 2004;39:1536-1569.

316
 Investigation Of The Fading Time Effects On
Microstructure And Mechanical Properties In
Vermicular Cast Iron

Mehmet Ekici1, Ugur Ozsarac1

Abstract
In this study, the fading time affecting the mechanical properties and microstructures of vermicular cast iron were studied.
Pig iron and steel scrap weighing about 12 kg were charged into the high frequency induction furnace crucible and
completely melted for production of vermicular cast iron. The slag was skimmed using a common flux. After fading time
was set at 1.3 and 5 minutes. In this way three vermicular cast iron was produced that same composition but different
phase structures. The microstructure of specimens was investigated and uni-axial tensile test and the charpy impact test
were performed and their micro-hardness measurements were done in order to characterize the mechanical behaviours of
vermicular cast iron.

Keywords: Vermicular cast iron, fading time, hardness, tensile test, impact test

1. INTRODUCTION
Vermicular cast iron is a promising engineering material containing graphite precipitates having a short, stubby, compacted
shape with rounded edges in a matrix basically similar to steel [1]. Vermicular cast iron possesses distinctive mechanical and
physical properties which are intermediate between those of grey cast iron and ductile cast iron. Vermicular cast iron
provides at least 70% higher tensile strength, 35% higher elastic modulus and approximately double the fatigue strength of
conventional grey cast iron [2]. Because of these attributes and its good thermal shock resistance, vermicular cast iron has
found application in the manufacturing of automotive engines, machine parts and rolls for various applications [3].
Fade, in ductile irons, is considered to be the loss of nodularity with time [4]. The concept of ‗fade‘ in ductile iron includes
the processes of magnesium fade as well as inoculant fade. These two terms both refer to the deterioration of microstructure
‗loss of nodularity‘ that accompanies the holding of treated and inoculated metal in the pouring vessel for some time prior to
actual pouring. The loss of residual magnesium (magnesium fade) results in the reversion of spheroidal graphite to
compacted or even flake graphite. This condition has been depicted graphically in studies [5]. As a result, foundries that
feature manual pouring almost always impose a time limit on how long iron can remain in the ladle prior to pouring. This
time limit is often seven to twelve minutes and its existence assumes that the iron will have lost sufficient magnesium content
after such an interval as to require pigging of the metal. The caution that accompanies such a procedure is the natural result of
the fear of producing a casting that fails in the field.

2. EXPERIMENTAL STUDIES
In this study, materials medium frequency induction furnace were used to produce vermicular cast iron with a capacity of
1800 kg. The materials used in this work were pig iron, steel scrap, ferro-silicon and a nodulizer. The chemical composition
of these materials is shown in Table 1. An easy way to comply with the symposium paper formatting requirements is to use
this document as a template and simply type your text into it.

1
Corresponding author: Yalova University, Vocational School of Yalova, Yalova, [email protected]

317
 Table 1. The chemical composition of alloy additions

Item C Si Mn Mg P Cu Ni Cr

Pig iron 4.15 0.85 0.06 0.01 0.04 0.02 - -

Steel scrap 0.10 0.19 0.66 0.01 0.02 0.28 0.09 0.05

Ferro - silikon 0.5 75 0.2 - - - - -

Nodulizer 0.2 45 - 5.5 - - - -

Pig iron and steel scrap weighing about 12 kg were charged into the high frequency induction furnace crucible and
completely melted. The slag was skimmed using a common flux. The molten cast iron was held at an elevated temperature
and some of the melt was poured into the melt holder connected with the carbon equivalent analyzer. Thereafter, the chemical
composition was adjusted and checked using the carbon equivalent analyzer. Once the main alloy additions were complete,
chilled samples were collected from each melt for chemical analysis before adding the nodulizer. Liquid treatment, using a
Fe–Si–Mg nodulizer, was performed by either directly injecting the nodulizer into the melt within the pouring ladle or by
placing the nodulizer on the bottom of ladle and then pouring the melt on to it. The tapping temperature ranged from 1350 to
1450 ◦C. The molten metal was poured into the carbon equivalent analyzer crucible to estimate the CE, after adding the
nodulizer. Vermicular cast iron with chemical compositions listed in Table 2 was prepared. The carbon equivalent (CE) value
was defined by:
CE = mass % C + (1/ 3 mass % Si + mass % P)

Table 2. Chemical composition of the vermicular cast iron sample (values in wt%)
Material C Si Mn P S Cr Cu Ti Mg Ces

VGI 3,71 2,13 0,50 0,02 0,01 0,02 0,39 0,01 0,01 4.44

3. RESULTS AND DISCUSSION

Figure 1 shows a series of micrographs obtained using a light microscopy. Figure 1a and f shows typical microstructures
obtained from the set at 1, 3 and 5 minutes. From Figure 1e it can be seen that the specimen exhibits a good homogeneous
graphite structure with a nodularity of less than 5%. Figure 1d and f shows the micrograph revealing the matrix structure
obtained following application of a nital (3%HNO3 in methanol solution) etch for 1, 3 and 5 minutes. From the micrograph it
can be seen that the VGI was present in a ferritic–pearlitic matrix containing approximately 22% ferrite and 78% pearlite.

318
 Figure 1. Micrographs of the produced vermicular cast iron with different set at 1, 3 and 5 minutes (a,b and c: non-etched and d,e and f:
etched).

Figure 2 shows amount of graphite obtained using a microstructure image analysis measurement program. The residual
magnesium was found to have the strongest influence on graphite. The effect of the residual Mg% on the compact graphite in
percent for the different set at 1, 3 and 5 minutes is shown in Figure 2. It can be seen that the compact graphite increases
approximately linearly with decrease residual Mg%. In this study, the obtained from the 3.71% C of vermicular cast iron
samples. Three fading time levels of graphite nodularity in a ferritic pearlitic matrix were obtained in Figure 3.

Figure 2. Amount of graphite observed in vermicular cast iron samples

Figure 3. Amount of phases observed in vermicular cast iron samples

The effect of the residual Mg% on the UTS and hardness for the samples with fading times of 1, 3 and 5 minute is shown in
Figure 4, respectively. Figure 4 indicates that the ultimate tensile strength increased with increasing residual Mg content for
all fading times, due to the increase in the campact graphite. From the tensile test results, an increment in yield and ultimate
tensile strength as 63 MPa and 26 MPa, respectively was observed in the fading time 3 minute. However, a significant
reduction has been observed in per cent elongation proving the decrease in toughness. From Figure 4, it can be seen that the
hardness was decreased by increasing the residual Mg and graphite, because after casting, not much more evaporation can be
going on.

319
 Figure 4. Tensile test and hardness results of vermicular cast iron specimens

As shown in Figure 5, notch impact test was applied to the samples at different fading time. In between 60°C and -30°C
temperatures, both samples‘ breaking energies has not been much change. However, the temperature of 0°C fracture energies
of the samples is increased. Fracture energy of 3 minute was observed to be better. In general, it can be said that the fracture
energies for each sample by looking at the work areas are generally between 0°C to +60°C to determine.

Figure 6. Impact test results of samples

4. CONCLUSION
This study was carried out to examine the fading times affecting the production of compacted graphite iron. This study shows
that microstructure has been found to occur of ferrite, pearlite and compact graphite, when the scope of work vermicular cast
iron castings is analysed. In conditions of spilled samples volume fraction of ferrite increases with increasing; fading times.
The tensile strength and hardness were increased by increasing the residual Mg%. In general, with decreasing of the amount
of pearlite in the internal structure of vermicular cast iron materials, the hardness values are decreasing. According to the
results obtained from the fracture energy, determination of vermicular cast iron material usage is recommended that at least at
room temperature.

5. REFERENCES
[1]. Y.-H. Shy, C.-H. Hsu, S.-C. Lee, C.-Y. Hou, Materials Science and Engineering A 278, 54-60, (2000)
[2]. G. Cueva, A. Sinatora, W.L. Guesser, A.P. Tschiptschin, Wear 255,1256–1260,(2003)
[3]. G.F. Geier,W. Bauer, B.J. McKay, P. Schumacher, Materials Science and Engineering A 413–414 339–345, (2005)
[4]. R. Gundlach, C. Loper, M. Bernardo, Composition of Ductile Irons, Ductile Iron Handbook, 87, (1992)
[5]. T. Frush, Y. Lerner, M. Fahmy, Inoculants Selection for Counter-Gravity Casting of Thin Wall Ductile Iron, International Inoculation
Conference Proceeding, (1998)S. M. Metev and V. P. Veiko, Laser Assisted Microtechnology, 2nd ed., R. M. Osgood, Jr., Ed. Berlin,
Germany: Springer-Verlag, 1998.

320
 Design of an Intelligent Control System to Prevent
the Ferroresonance Effect in Measuring
Transformers
Hilmi Zenk1

Abstract
Voltage and current transformers used for measurement and protection are one of the most used elements of the power
system. The B-H characteristic of the transformers provides electromagnetic energy conversion in a linear region with a
high efficiency. With the transformer's ferromagnetic core saturating, the energy conversion efficiency begins to fall.
Nonlinear inductance of the transformer and elements such as resistance and capacitance in the nonlinear loads of the
power system connected to the current transformer has the potential to interfere suddenly and cause a ferroresonation
phenomenon. In this study, it has been determined that transformers, especially those operating in high voltage systems and
compensation plants, produce destructive electrical parameters in the system due to the fact that they work together with
the continuous closing-opening switches. It has been determined that it is possible to remove randomly operated switch-off
openings which cause this destructive condition by means of a control system operating at appropriate times using
semiconductor switches.
Keywords: Current transformer, voltage transformer, ferroresonance, switch position control, power systems

1. INTRODUCTION
Transformers are one of the most commonly used elements in power systems. While transformers operate in the linear
region, they provide electromagnetic energy conversion with high efficiency. However, as the transformer's ferromagnetic
core begins to saturate, the energy conversion begins to fall and the losses begin to increase. Reduction of iron losses of
electric machines, alternative magnetization of ferromagnetic materials, iron losses of dynamo and transformer plates,
hysteresis losses in static and symmetric magnetism, reduction of foucault losses have been the main aims of researchers for a
long time[1].
The ferroresonance phenomenon is an electrical problem with the expansion of the electrical energy system, which causes
complex, high-voltage sudden emergence and high-level harmonic distortion. Ferroresonance is different from the
resonances occurring in linear systems. While resonance in linear systems creates sinusoidal current and voltage waveforms
with high amplitude, ferroresonance causes irregular or complex (chaotic) wave forms in high amplitude current and voltage
values [2].

1.1. Ferroresonance Events

Ferroresonance is an unstable operation in the system consisting of resistance, capacitance and inductance. By changing the
value of one of the electrical elements in the circuit, there is a sudden increase in the current and voltage values between the
ends of the other elements. Ferroresonance may be more common in future electrical systems. Increasing the transmission
and distribution voltages will create a more varied relationship between the line capacitance and the magnetic saturation
curves of the transformers today. This change is due to an increase in the likelihood of ferroresonance [3].
Transformers are at the top of the elements commonly used in the transmission and distribution of electricity. The level of
electromagnetic power conversion that these transformers will make depends on many factors, but one of the most important
ones is the magnetic permeability of the magnetic core of the transformer. In this case, the magnetic core constant value can
be thought of as a varying inductance. It is also evident that the loads connected to the output of the transformer are
composed of elements such as resistance, inductance and capacitance. The necessary conditions for the resonance
phenomenon seen in the electric circuits are thus formed. The ferroresonance phenomenon can be encountered in systems
containing all kinds of transformers. It is difficult to predict how the electrical parameters will behave in resonant elements.
In the ferroresonance case it is highly unlikely to predict how additional loads on the system, which increase the level of
resonance and cause it to switch to chaotic states, or the random removal of some of the loads from the system, will affect
this situation.

1
Corresponding author: Giresun University, Department of Electrical-Electronics Engineering, 28200, Güre Campus Giresun, Turkey.
[email protected]

321
 + VC - m
I
+
C
Vs ~ R VL L
S
-
m

(1)
C capacitor, supply voltage and frequency are affected. Source voltage,

(2)

is defined by the expression. Duffing, resonance, sub-and super resonance of the first researchers who work on issues and
evidence given by equation (3) can be likened to the Duffing oscillator [4-7]. In equations "n" indicates the degree of
expression magnetization.

(3)

The other parameters of the equation, k, a1, b1 and G equations (4-7), respectively.

(4)

(5)

(6)

(7)

2. MEASURING TRANSFORMERS
Measuring high currents and voltages directly with measuring instruments is both difficult and dangerous in terms of safety.
The directing of high currents through the ammeter requires the ampermeter to be physically very large. Measuring high
voltages directly with a voltmeter can be dangerous for safety reasons. For this purpose, measurement transformers are used
to measure high voltage and large currents. Measuring transformers reduce the large current and high voltage to be measured
to a level that can be safely measured by measuring instruments. It is connected to the high voltage or large current circuit
which will measure the primary voltage of the measuring transformers. Ampermeter, voltmeter, wattmeter, various relays or
control circuits are connected to the circuit of the secondary circuits.

2.1. Voltage Transformers

Transformers that transfer the primary circuit voltage to the secondary circuit elements by shrinking it within the conversion
ratio and the phase difference between the primary and secondary winding voltages is about zero degree. It also provides
insulation of the measuring and protection devices from the high voltage system.

Figure 2. (a) Voltage transformers (b) Current transformers

322
2.2. Current Transformers
They are called special transformer current transformers that reduce the current to which they are connected in the desired
direction and supply the devices connected to the secondary terminals with this current and isolate them from the high
voltage.

3. MATLAB / SIMULINK DESIGN OF THE WHOLE SYSTEM

The response of the system in different closing and opening scenarios has been investigated by using the electronic angle
control system designed for the idle working and loaded working states of the transformers used for measurement in high
voltage systems. In Figure 3, the entire system used in simulation is given a Matlab/Simulink overview. After measuring the
source voltage, the inverter TR1 is connected to the coil P1 of the voltage transformer after the circuit breaker named
Breaker_1. This transformer P1 and P2 voltage data was measured and sent to sub-blocks MEASUREMENT and
CONTROL. In order to be able to create High Voltage conditions, the inverted-connected S1 terminal of the TR1 transformer
is connected to the P1 terminal of the rectifier TR2 transformer. The ferroresonance effect was generated at the ends of two
separate circuits S1 and S2 using loads according to the labeling power of the investigated TR2 transformer. The electrical
parameters of all elements in these two circuits have been sent to the MEASUREMENT subblock.

Figure 3. Matlab/Simulink overall view of the entire system used in simulation

3.1. Angle Control System (ACS)

The switch that performs the ON/OFF operation of the system plays a very important role in generating the ferroresonance
gain. Depending on the period of the voltage signal at the point where the designed switch is connected, it must be capable of
opening or closing at the desired electrical angle.
A subprogram is sent in the Matlab/Simulink simulation program that is designed as an example of the voltage signal taken
from the measured phase. Here, zero crossing point of the relevant phase is determined, and with the intelligent control
system prepared, it provides the opening and closing operation between 0 ° and 360 ° with the desired electrical sensitivity.
Figure 4. shows the logic circuit in which a sub-circuit that reads the zero transition of the sign to select the switching angle is
generated. The signal is held until the entered delay and the command is sent to the circuit breakers after the time has elapsed.
When the switch is closed for closure, the lower block designed to select the switching angle and the signal from the sensor
that measures the phase voltage of the respective phase. The incoming signal is made to be turned off after the delay of
loading the program. Later, these signals are detected from zero and the moments when the counter passes from zero are
recorded. When the trip signal is received and the counter is reset, the delay block will trip with so much delay if the delay is
applied. If switching on and off delays are requested, a variable is assigned and can be changed at any time in the program
with an mfile file.

323
 Figure 4. ACS calculation sub-block prepared in Matlab / Simulink simulation program
Transformer Primary

On sign On sign On sign On sign

Winding Voltage
TR1 Voltage

Off sign Off sign Off sign

Off sign Off sign
Signs of closing and opening
produced by the BADS1

ON ON ON
Switch

OFF OFF OFF OFF

Figure 5. The TR1 voltage transformer primary winding voltage to the input of the BADS1 switch and the ON and OFF signals generated at
the output of the BADS1 switch

Figures 5. And 6. show the closing and opening signals generated at the switch output which come to the angle selector
position switch as a result of the random ON and OFF operations performed by the user. When the graphs are examined in
detail, when the first closing sign arrives, the system starts to look similar to 0,027s, which corresponds to about 126º of the
full signal in degrees. BADS (ACS block name) arrives at this closing sign from the beginning of the new full period, ie 0.04.
Added the delay in degrees as much as the user wanted from the moment, ie it produced the closing mark in 0.0417s. Here
the delay, T: period duration (s), Tstart: period start time (s), TOFF: delayed shutdown time (s) produced by the BADS switch,
Tdelay: delay time (s), αdelay: The delay in degrees is expressed by equations (8) and (9).

(8)

(9)

When these equations are used it is seen that the delay is 30º. When x is examined, the closing times are 0,04 seconds, 0,12
seconds and 0,16 seconds after 30 seconds, 0,08 seconds, 0,14 seconds when no opening delay times are reached, It occurred
in 0.18 seconds.

324
 The first time zero crossing of the
voltage sign on the BADS1 switch ON signal time entering
BADS 1 switch

OFF signal time entering

Time of the closing signal BADS1 switch
entering the BADS1 switch

30 ° delayed ON signal
generated by BADS1 switch

0 ° delayed OFF signal

generated by BADS1 switch

Figure 6. Detailed analysis for a short time interval between the sign of the BADS-1 key input and its output

When the Figure 7. is examined, when the mains voltage of the TR1 voltage transformer primary winding is applied at the
level of 30º of the mains voltage of 220V, the signal has reached the normal level at 90º in the same period, but in the
following full voltage, the voltage exceeded about 400V. Similarly, when the state of the secondary winding is applied at the
level of 30º, the signal reaches its normal level at 90º of the same period, but the voltage following this period reaches about
40kV.

Figure 7 Some parameters of TR1 voltage transformer-time change curves

When examined in Figure 8, a cosine curve with a non-symmetric, peak value of 2 V.s, starting from zero at 30º, at which the
switch closes, is visible. The peak value in the unit value of the magnetic flux is at the level of 1,8575 pu, while the negative
alternation is at the level of -0,8915 pu. Also, when the BADS-1 key is first opened, the permanent flux in the TR1 traf is at -
0,725 pu. When the second closure comes, the magnetic flux value starts at -0,725 pu at which the switch is opened. The
peak values of the magnetization and excitation currents are at the 5A level.

325
 Figure 8. Detailed analysis for a short time interval between the sign of the BADS-1 key input and its output

4. RESULTS
The proposed system has shown that the ferro-resonance phenomenon is more likely to occur as the key-closing angle
increases, the current and voltage signals in the system become distorted, the harmonic values increase, and small internal
cycles are formed on the hysteresis curve. The total harmonic distortion at the moment of closing of the switch, which is
closed with large angle values, reaches very high values.
On the other hand, controlled switch-on is also performed at the same time as the switch designed in this study.In case of
controlled switching operation, it detects the moment when the phase parameter which will cut off the energy goes from zero
and opens the system with necessary delays. With this maneuvering, it is possible to reduce the influence of residual flux,
resonance and harmonics in the transformer core.
It can be predicted that very fast power electronic circuit elements to be designed in the future can be used extensively due to
the introduction of circuit breakers to eliminate high voltage and current problems, as well as to the contribution of controlled
switching in power distribution centers and cutter measuring cabinets.

REFERENCES
[1]. H. Zenk, A. S. Akpinar, ―High Voltage Instrument Transformers Ferroresonance Events Occurring in the Investigation of the Circuit
Breaker Switch Positions‖, Electrical and Electronic Engineering, vol. 4, no. 1, pp. 10-19, 2014.
[2]. E. Yigit, ―Guc Trafolarinda Ferrorezonans Olaylarin Kaotik Analizi‖, Yuksek Lisans Tezi, Fen Bilimleri Enstitusu, Sakarya
Universitesi, Haziran 2009.
[3]. Alparslan Y., ―Elektrik Sistemlerinde Manyetik Cekirdekteki Doyma Dolayisiyla Hasil Olan Gerilimler (Ferrorezonans)‖, Elektrik
Muhendisligi 6. Teknik Kongresi, 1983, Bildiriler Kitabi, pp.147-153.
[4]. C. Kocatepe, M. Uzunoglu, R. Yumurtaci, A. Karakas ve O. Arikan, Elektrik Tesislerinde Harmonikler, Birsen Yayinevi, 2003.
[5]. R. G. Kavasseri, ―Analytical Prediction of Subharmonic Oscillations in a Ferroresonant Circuit‖, Department of Electrical and
Computer Engineering, North Dakota State University, 58105 - 5285, Fargo, ND, USA.
[6]. G. W. Swift, ―An Analytical Approach to Ferroresonance‖, IEEE Transactions On Power Apparatus And Systems, 88, 1, January 1969.
[7]. H. Zenk, ―The Effect Of Turning-Off Angle On Ferroresonance Event‖, PhD. Thesis The Graduate School of Natural and Applied
Sciences, Karadeniz Technical University, 2014.

326
 An Investigation Of Electrical Response On
Interleaved Buck Converters Using Different Type
Current Control Methods
Hilmi Zenk1

Abstract
DC converters are widely used in industrial area. Rapidly advancing technology is increasing the expectation of more
efficient and smaller size power electronics systems. Adjusted DC voltages required by electrical loads are often provided
by DC converters. The coexistence of the same converter types brings advantages such as increased system durability and
fault tolerance due to current sharing capability. The use of interleaved switching signals in parallel-connected power
converters is one of the preferred methods because it reduces the overall system size by reducing filter element size. In this
study, the continuous transfer mode of the vine buck converter; The unit power factor (UPF) and the total harmonic
distortion (THD) of the input current are investigated using proportional-integral (PI) control and avarage sliding mode
control (ASMC). The Buck Converter simulates MATLAB / Simulink at 100W power and 20 kHz switching frequency and
analyzes the control methods. In the analyzes, the THD effect of the converter's response to the load changes in the control
methods and other parameter changes was investigated.
Keywords: proportional-integral (PI) control; avarage sliding mode control (ASMC); unit power factor (UPF); total
harmonic distortion (THD).

1. INTRODUCTION
The widespread use of semiconductor switches has accelerated the pace of development of the technology, which has brought
with it some electrical problems. At the beginning of these problems, the harmonics in the electricity distribution system
increase, and the sinusoidal current waveform deteriorates. Many researchers are working on different types of circuit designs
to reduce the harmonics in the currents feeding the systems that contain the power switches.
The power sources used in power electronic converters draw non-sinusoidal and harmonic currents from the network. These
currents drawn from the system cause low power factor (PF) and high line currents to be generated. Decreasing the quality of
the current drawn from the network; Resulting in deterioration of the mains voltage, increased losses, the emergence of
electromagnetic parasites and inefficient use of consumed power [1]. Many researchers are working on different types of
circuit designs to meet the electrical quality standards introduced by some international organizations and to improve the
power coefficient.Based on these standards, using passive circuits, techniques such as reducing harmonic components,
correcting active power coefficient and active rectifiers are used to derive electrical adverse effects [2]. There are different
circuit topologies of power converters for correcting the power factor such as buck, boost, flyback, interleaved structure [3-
7].
Most DC/DC converters are usually provided a low output voltage and high output current to load. Therefore, a conventional
interleaved buck converter (IBC) as shown in Figure 1 is widely adopted, because it has a simple structure, high output
current density and low output current ripple. However, in high step-down voltage applications, it suffers from extremely
short duty ratio and high component stresses, resulting in low conversion efficiency [8-11].

2. INTERLEAVED BUCK CONVERTER

The circuit diagram of IBC is given as shown in Figure 1. This is equivalent to a parallel combination of two sets of switches,
diodes and inductors connected to a common filter capacitor and load [12]. The switches are operated out of phase. Assume
the converter operates with duty ratio less than 50% and in continuous conduction mode [13].
Figures and tables must be centered in the column. Graphics may be full color.

1
Corresponding author: Giresun University, Department of Electrical-Electronics Engineering, 28200, Güre Campus Giresun, Turkey.
[email protected]

327
 L1
Q1 iL1

+ VL1 - iL
D1
+ +
L2
Q2 iL2
Vs Vin Cout Vout RL
D2
- -

Figure 55. A circuit diagram of 2 phase IBC

A two phase IBC will operate in four different modes and is explained as follows:

2.1. IBC Operation Mode-1

In Operation mode-1 switch Q1 is turned on by giving a gate pulse. At the same time switch Q2 is off. Current flows through
the switch Q1, inductor L1 and load, making current through L1 to increase as long as Q1 is turned on. During this time current
in L2 decreases linearly. The equivalent circuit is as in Figure 2. The variations of iL1 and iL2 during T are given by,

(1)

(2)

L1
Q1 iL1

+ VL1 - iL
D1
+ +
L2
Q2 iL2
Vs Vin Cout Vout RL
D2
- -

Figure 2. Equivalent Circuit of IBC in Mode 1

2.2. IBC Operation Mode-2

Mode 2: Since IBC operates with a duty cycle less than 0.5, in this mode both the switches are OFF. Diodes D 1 and D2 are
the conducting devices. The equivalent circuit is illustrated in Figure 3.The energies stored in L 1 and L2 are released to the
load through the forward biased diodes. So iL1 and iL2 are decreased linearly. Thus the variations in iL1 and iL2during T2 are
given by,

(3)

(4)

L1
Q1 iL1

+ VL1 - iL
D1
+ +
L2
Q2 iL2
Vs Vin Cout Vout RL
D2
- -

Figure 3. Equivalent Circuit of IBC in Mode 2

2.3. IBC Operation Mode-3

During T3 Q2 is turned On and Q1 turned off. The equivalent circuit is illustrated in Figure 4. The turning on of Q 2 charges the
inductor L2 and since Q1 is off inductor L1 is discharged to the load. The variations in iL1 and iL2 during T3 are given by,

(5)

328
 (6)

Equation (5) shows iL1 linearly decreasing during T3 since the slope is negative and a constant.
L1
Q1 iL1

+ VL1 - iL
D1
+ +
L2
Q2 iL2
Vs Vin Cout Vout RL
D2
- -

Figure 4. Equivalent Circuit of IBC in Mode 3

2.4. IBC Operation Mode-4

The operating mode is same as mode 2. The variations in iL1 and iL2 during T4 are given as in equation (3) and (4).

Q1
ON ON

Q2 ON ON
İL1

İL2

İL

T4 T1 T2 T3 T4 T=T1+T2+T3+T4 t

Figure 5. Inductor Current Waveforms

3. CURRENT CONTROL METHODS

The only difference in control methods for interleaved converters is the number of feedback cycles. For the N-phase
converter, there is N number of feedback cycles. Additional circuitry is needed to enable phase shift operation in the
interleaved loop. Hence, all control techniques are designed according to single phase and D duty ratio is shifted by phase
difference. This section describes control methods that synchronize the input current with the basic component of the input
voltage [14].

3.1. Proportional-Integral Control Method

Proportional-integral (PI) controller is one of the commonly used techniques to correct the power factor. In this method, the
average inductor current is taken and this average value is controlled by switching.

(7)

(8)

(9)

(10)

In equations, VEAV: Proportional-integral controlled voltage error amplifier, VEAI: Proportional-integral controlled current
error amplifier, D: duty cycle. In the proportional-integral controller, Kp and Ki can be selected according to the method of

329
Ziegler-Nichols. The PI control method follows the actual current reference current to provide a unit power factor and low
THD [14]. Figure 6 shows the proportional integral control method.

Figure 6. PI controller block for Simulink Model of IBC

3.2. Average Sliding Control Method

Control of systems with nonlinear and complex dynamics is very difficult with classical supervisory methods. The Average
Sliding Control Method (ASCM) can be an effective control method in the control of such systems. The purpose of applying
the ASCM method to closed-loop control systems is to push the error to the sliding surface or, alternatively, to the switching
surface and keep it at that surface. Since the slip surface is defined as a function which is a linear combination of state
variables, the state variables are linearly dependent on this surface. In this case, the system is reduced to a level independent
input variable, and the system is controlled by a reduced control rule [15].

Figure 7. Voltage controller block for Simulink Model of IBC

The voltage controller has the same structure as shown in Figure 7 according to all control methods. Figure 8 shows the
average sliding mode control method.

Figure 8. Average sliding current controller for Simulink Model of IBC

4. SIMULATION
Simulation studies were performed using the MATLAB / Simulink program. Selected parameters in simulation studies are
given in Table 1. The two-cell interleaved buck converter is shown in Figure 9.
Table 29. IBC Simulation Parameters

330
 Parameters Symbol Value

Number of Phases Nphase 2

Input Voltage Vin 150-200 V

Output Voltage Vout 24 V

Switching Frequency F 65 kHz

Per phase Ripple current %10

ΔI
Output Current Iout 5A

Indutors per phase L1, L2 100 μH

Output voltage ripple Vo_ripple ≤ 20 mV

Output Capacitor Cout 22 μF

Figure 9. Simulink Model of IBC

Figure10. (a) In PI-controlled IBC, input voltage Vin, output voltage Vout and output current Iout (b) Detail in 5ms

331
 Figure 11. (a) Change in inductance currents of L1 and L2 (b) Detailed analysis of changes in currents of L1 and L2

Figure12. (a) In ASMC- controlled IBC, input voltage Vin, output voltage Vout and output current Iout (b) Detail in 5ms

Figure 13. (a) Change in inductance currents of L1 and L2 (b) Detailed analysis of changes in currents of L1 and L2

For the described control methods, the switching frequency of the power electronic transducer is taken as 65 kHz. Results
according to control techniques are presented in Table 2 below.
Table 2. Simulation Results

Simulation Number 1 2 3 4 5 6 7 8

Power (W) 100 100 100 100 50 100 100 100

L1=L2 (mH) 100 100 100 100 80 50 50 50

Cout (μF) 22 22 22 22 22 22 22 22

Vout (rms) 200 100 250 220 200 200 150 100

Proportional-Integral Control Method

332
 Power Factor (PF) 0,99 0,99 0,99 0,99 0,99 0,99 0,99 0,99

THD (%) 5,1 2,0 6,3 4,7 5,2 23,1 8,5 4,5

Average Sliding Control Method

Power Factor (PF) 0,99 1 0,99 0,99 0,99 0,99 0,99 0,99

THD (%) 4,8 1,4 5,4 3,9 4,2 15,2 6,3 3,9

5. RESULTS
In this study, proportional-integral and average sliding mode network current control methods are applied to a two-cell buck-
type power factor correcting circuit and simulated in Matlab / Simulink program. The results obtained are very close to each
other. The control techniques used have been based on practical work using fixed switching frequency. It has been observed
that both of the tested control techniques exhibit a structure resistant to output load changes, as can be seen from the graphs.
It has been observed that the proportional-integral control technique in the designed 100W simulation circuit is more
sensitive to inductor variations than the average floating mode control technique. It has been observed that when the inductor
value is reduced, the proportional-integral controller coefficient adjustments must be made again and the average sliding
mode method maintains its performance. Capacitor change affects output voltage fluctuation

REFERENCES
[1]. Karik F., ―Ortalama Kayan Kip Metodu ile Denetlenen Iki Fazli Sarmasik Yapili Yukseltici Tip Donusturucunun Performans Analizi‖,
Elektrik, Elektronik, Bilgisayar, Biyomedikal Muhendisligi Bilimsel Dergisi, no.4, vol.2, 2012.
[2]. Chen, Z., Raymond, B.R. ve Fred, C.L., "Design Analysis of a Hysteresis Boost Power Factor Correction Circuit", Power Electronics
Specialists Conference PESC '90 Record-21st Annual IEEE, pp.800-807, 1990.
[3]. Qiao, C., ve Smedley, K.M.,‖A topology survey of single-stage power factor corrector with a boost type input-current-shaper‖, IEEE
Transactions on Power Electronics, vol.16, no.3, pp.360 – 368, 2001.
[4]. Po-Wa, L., Yim-Shu, L., Cheng, D.K.W., ve Xiu-Cheng, L., ―Steady-state analysis of an interleaved converter with coupled
inductors‖, Industrial Electronics IEEE Transactions, vol.47, pp.4, pp.787 – 795, 2000.
[5]. Yao-Ching, H., Te-Chin, H., ve Hau-Chen, Y., ―An Interleaved Boost Converter with Zero-Voltage Transition‖, IEEE Transactions on
Power Electronics, vol.24, no.4, pp.973 – 978, 2009.
[6]. Alonso, J.M., Dalla Costa, M.A., ve Ordiz, C., ―Integrated Buck-Flyback Converter as a High-Power-Factor Off-Line Power Supply‖,
Industrial Electronics IEEE Transactions, vol.55, no.3, pp.1090 – 1100, 2008.
[7]. Jun, Z., Lu, D.D.-C., ve Ting, S., ―Flyback-Based Single-Stage Power-Factor-Correction Scheme with Time-Multiplexing Control‖,
Industrial Electronics IEEE Transactions, vol.57, no.3, pp. 1041 – 1049, 2010.
[8]. Agarwal, V., Aggarwal, R.K., Patidar, P., Patki, C., ―A novel scheme for rapid tracking of maximum power point in wind energy
generation systems‖, IEEE Trans. Power Electron. vol.25, pp.228–236, 2010.
[9]. Wong, P., Xu, P., Lee, F.C., ―Performance improvements of interleaving VRMs with coupling inductors‖, IEEE Trans. Power
Electron., vol.16, pp.499–507, 2001.
[10]. Chen, Y.M.; Tseng, S.Y.; Tsai, C.T.; Wu, T.F., ―Interleaved buck converters with a single-capacitor turn-off snubber‖, IEEE Trans.
Aerosp. Electron. Syst., 2004, 40, 954–967.
[11]. Munoze, ―C. Study of a New Passive Lossless Turn-Off Snubber‖, In Proceedings of the International Power Electronics Congress,
Seoul, Korea, 26–31 October 1998, pp. 147–152.
[12]. Dodi Garinto, ―A Novel Multiphase Multi-Interleaving Buck Converters for Future Microprocessors‖, 12thInternational Power
Electronics and Motion Control Conference, PEMC 2006, pp. 82-87 , Aug-sep 2006.
[13]. Joseph A., Francis J., ―Design and Simulation of Two Phase Interleaved Buck Converter‖, International Journal of Advanced Research
in Electrical, Electronics and Instrumentation Engineering, vol.4, Special Issue 1, 2015.
[14]. Karik F., Iskender I., Karaarslan A., Genc N., ―Sarmasik Yapili Tek-Faz Dogrultucunun Farkli Akim Kontrol Yontemleriyle
Performans Analizi‖, Journal of the Faculty of Engineering and Architecture of Gazi University, vol.29, no.3, pp.443-450, 2014.
[15]. Kayisli, K.., Tuncer, S., Poyraz, M., ―Kayma Mod Denetleyici Kullanilarak Aktif Guc Faktoru Duzeltimi‖, Pamukkale Universitesi
Muhendislik Bilimleri Dergisi, vol.14, no.3, pp.253-260, 2008.

333
 Design of IPM Synchronous Motor for Gearless
Elevator Applications
Hicret Yetiş1, Erkan Meşe2

Abstract
Interior permanent magnet synchronous motors (IPMSM) have been commonly used to meet challenging demands of high
performance industrial applications. IPMSM have several advantages such as elimination of rotor copper loss, high power
density and efficiency, high starting torque. In this paper, the design of IPMSM for gearless elevator has been investigated.
Gearless elevator systems driven by PMSM have many advantages compared to traditional asynchronous motor
applications. Due to highly sensitive absolute encoder coupled to motor and closed-loop driver system providing sensitive
control, PMSM based gearless elevators show high performance at lifting up, stopping and moving. However, drawbacks
of IPM machines are high torque ripple and rising cost of rare-earth permanent magnets such as NdFeB and SmCo. The
variation of magnetic reluctance between the flux barriers and teeth causes the torque ripple. In this study, different
slot/pole combinations and number of flux-barrier layer have been investigated to reduce torque ripple. Furthermore whole
design is based on ferrite magnet which is cheaper and more common. MotorSolve BLDC Finite Element Analysis based
software has been used. Eventually attributes such as low cost, high efficiency, high torque density and lowest possible
torque ripple have been achieved for the design of IPMSM in gearless elevator applications.

Keywords: interior permanent magnet (IPM), machine design, gearless elevator application, torque ripple, ferrite
magnet.

INTRODUCTION
Permanent-Magnet (PM) machines are widely demanded in industrial applications owing to their high torque density
and efficiency. Both surface-mounted PM (SPM) and interior PM (IPM) rotor types are used depending on the application
[1], [2].IPMsynchronous motors(IPMSM)offer a wide constant power speed range than SPM synchronous motors. For this
reason IPSM widely preferablein traction application.
The IPMSMs may be very attractive candidate for direct driven gearless elevator systems owing to its ability to show
high performance at lifting up, stopping and moving. Beside comfort, one of the most important features of gearless elevator
systems driven by PMSM is providing energy conservation compared to systems designed with traditional asynchronous
motor and gear box[3].Gearless elevators can be designed with smaller electric motors for the same weight capacity and the
reduction can be as high as 50 % [4].
Gradually, there was the convenience of adopting PM with high magnetic energy (i.e., rare-earth PMs, as NdFeB
magnets) so as to increase the flux density and to reduce the dimension of the machine and the associated cost [5]. On the
other hand, in the last years, the cost of rare-earth PMs such as NdFeB and SmCo is significantly increased and the motor
manufacturer reconsidered the opportunity to use cheaper PMs, such as Ferrite PMs.
This paper presents design of ferrite based IPM motor for direct drive gearless elevator applications. First of all,
elevator system requirements (for an eight people carrying capacity) are determined.Then the motor is designed based on
these requirements. Additionally,as in the approach presented in [6] different slot/pole combinations and number of flux-
barrier layer have been investigated to achieve an optimum IPMSM design.

REQUIREMENTS OF ELEVATOR SYSTEMS

The significant requirements in the motor design of gearless elevator systems are torque and speed. These two
parameters can be calculated by operating speed, cabin weight capacity, type of suspension and pulley diameter of the
designed elevator system[7].

1
Corresponding author: Yildiz Technical University, Department of Electrical Engineering, 34220, Esenler/İstanbul, Turkey.
[email protected]
2
Corresponding author: Ege University, Department of Electrical and Electronics Engineering,35100, Bornova/İzmir, Turkey.
[email protected]

334
 In an elevator system with 675 kg weight (for 8 people) and 1 m/s cabin speed, the system requirements can be
calculated. In the calculation, it is assumed that counterbalance weight is selected so that it is equal to the, total ofweights of
entire cabin and half of the load.
(1)

Where,
: Rated Torque ( )

: The radius of drive pulley ( ), (0.12)

: The force of gravity , (9.88)

: Load weight ( ), (675)

: Cabin weight, : Balance weight ( )

Assumed,

u: The coefficient for suspension type. 1 for direct suspension, 2 for 2:1 suspension. Design was carried out for u=2.
: Well and rope system efficiency (80%)

= 250

Motor rated speed;

( ) (2)

v: cabin vertical velocity = 1

, 16.75

( ) (3)

, .

Rated power;
,( ) (4)

, .

So following motor parameters are fixed before starting design procedure,

, , .

THE FERRITE BASED IPM MOTOR

Interior permanent magnet synchronous motors (IPMSM) have been commonly used to meet challenging demands of
high performance industrial applications. IPMSM have several advantages such as elimination of rotor copper loss, high
power density and efficiency, high starting torque.However, drawbacks of IPM machines are high torque ripple and rising
cost of rare-earth permanent magnets such as NdFeB and SmCo. The use of rare-earth materials, which is an important aspect
of the highperformance PM motor, should be reduced because of the high cost and the unpredictability involved in procuring
such materials [8].
Most of up-to-date PM synchronous machines are based on rare-earth magnetic materials, namely, NdFeB grades,
because of their large remenance and coercivity values [9].Besides, cheaper PMs, such as ferrite-based magnets, are still
applicable even though exhibiting poor magnetic properties when compared with other rare-earth materials. As the prices of
NdFeB PMs has increased rapidly and motor manufacturers re-evaluated the opportunity to use cheaper PMs, such as Ferrite
PMs.
Because of these reasons, Ferrite magnetwhich is cheaper and more common has been used to reduce cost in this study.

DESIGN CONSIDERATIONS
IPMSMs are vulnerable to producing significant amounts of currentinduced torque ripple whose amplitude can vary
significantly depending on the operating conditions [10].Hence, the most significant design consideration for IPMSM is the
torque ripple, whichis generated as a undesirable by-product ofthe interaction between rotor and stator MMFwaveform. This
situationis intolerable in the most of applications.

335
 The main purpose in this study is to reduce the torque ripple toat minimum level. For this purpose, three different
IPMSM have been designed to reach the optimum design with the power, speed and torque determined in the section 2.

Table 1. Specifications for the IPM Machines

Item Value (Unit)

Rated power 4.2 kW

Rated speed 160 rpm

Rated torque 250 Nm

Rated current 13 A

Stator outer diameter 250 mm

The specifications of three Rotor

IPMSMsouter under
diameterinvestigation are shownin
149 mm Table 1.The machines are designed to
investigate the effect of slot/pole combinations and numbers of flux-barrier layer on the torque ripple are summarized in the
Table 2. Stack length 200 mm

Table 2. Summary
Air gap length of designed IPMSMs 0.5 mm

PM material Ferrite (Y40)

IPM name Stator slots number Flux barrier number Rotor pole number
Coil fill factor % 40
IPM1 33 3 8

IPM2 33 2 8
(a) IPM1 (b) IPM2 (c) IPM3
IPM3 42 3 8

Figure 1. Stator and rotor geometries of the IPM machines.

Fig. 1 shows the stator and rotor geometries of the three machines under study. For all motors, the rotor poles are
designed ferrite PMs. The parameters of IPMSMs should be optimized to achieve the minimum torque ripple at the same
operating speed and under the same thermal and volumetric constraints.Furthermore, three machines are also analyzed with
MotorSolve BLDC Finite Element Analysis based software.

ANALYSIS RESULTS

o Influence of Slot/Pole Number Combinations

In this section to investigate the influence of torque ripple on the number of slot/pole combination, IPM1 with 33-
slot/8-pole and IPM3 with 42-slot/8-pole are analyzed.Two IPMSM with different stator slot number but same pole number
and rotor structure are shown in Fig.1 (a, c). Average torque and torque ripple analysis results are shown in Fig. 2 and Fig. 3.
As can be seen from the analysis results, the slot/pole combination is considerably effective on the average torque and torque
ripple.

336
Figure 2. Average Torque (IPM1, IPM3) Figure3. Torque Ripple (IPM1,IPM3)

o Influence of Flux Barrier Layers

In order to investigate the effect of flux barrier number on average torque and torque ripple, two IPMSM rotors with
different flux-barrier number but the same stator structure and pole numbers are analyzed. Analysis results of IPM1 with 33-
slot/3-layer flux-barrier and IPM2 with 33-slot/2-layer flux-barrier are shown in Fig. 4 and Fig. 5.From analysis results, it is
observed that the multi-layer flux-barrier design has significant impact on motor performance.

Figure 4. Average Torque (IPM1, IPM2) Figure 5. Torque Ripple (IPM1, IPM2)

o Flux Distribution

In this sectionflux distribution maps for two machines are shown in Fig. 6.
( a ) IPM1 (b) IPM2 (c) IPM3
Figure 6. Flux distributions of the IPM machines

o Efficiency
The efficiency levels of the three machines designed to investigate the influence of slot/pole combinations and
number of flux-barrier layer on torque ripple are shown in Fig. 7.

337
 Figure 5. Efficiency

o Material Consumption and Cost

When machines are compared in terms of their active material weight, it turns out to be steel, copper and magnet
consumptions are comparable. Material consumptions of the machines are summarized in the
Table 3. Summary of material consumptions

Weight (kg)
Component
IPM1 IPM2 IPM3

Rotor core mass 9.88 9.04 9.88

Rotor magnet mass 3.62 4.01 3.62

Stator core mass 26.5 26.5 19.17

Stator winding mass 8.19 8.19 13.1

Total mass 48.19 47.74 46.3

Table 4. Summary of material cost

Price (USD)
Name
IPM1 IPM2 IPM3
Rotor core steel (0.8 USD/kg) 7.904 7.232 7.904
Permanent Magnet (7 USD/kg) 25.34 28.07 25.34
Stator core steel (0.8 USD/kg) 21.2 21.2 15.76
Stator winding steel (7 USD/kg) 57.33 57.33 91.7
Total cost 111.774 USD 113.832 USD 140.704 USD

Table 3.In accordance with Table 3. material cost of three machines are calculated. Calculated values are shown
Table 4.

CONCLUSIONS
In this paper, ferrite based three IPM machines are presented. Targeted application is gearless elevator system where
low torque ripple is desired. Elevator system requirements are determined andbased on these requirements motors are
designed. These machines are analyzed to investigate the effect of slot/pole combinations and number of flux-barrier layer on
the torque ripple. Also, the flux distribution, efficiency and material consumption and costof the three machines are
presented.Analysis results show that IPM1 with 33-slot/8-pole is superior to the other IPM machines in several aspects.

338