International Journal of Power Electronics and Drive Systems (IJPEDS)

Vol. 13, No. 4, December 2022, pp. 1951~1965

ISSN: 2088-8694, DOI: 10.11591/ijpeds.v13.i4.pp1951-1965  1951

Analysis of SiC and SiC-Cascode MOSFET in the design of

power electronics converters for more electric aircrafts

Osmar Felipe Alves Eleodoro, Vitor Fonseca Barbosa, Gustavo B. Lima, Luiz Carlos Gomes Freitas
Faculty of Electrical Engineering, Federal University of Uberlandia, Uberlândia, Brazil

Article Info ABSTRACT

Article history: Wideband gap (WBG) semiconductors have developed rapidly in recent
years, enabling greater efficiency and power density in the design of power
Received Jun 25, 2022 electronics converters for several areas of application. In this sense, this
Revised Aug 11, 2022 work analyzes the efficiency and specificities of silicon carbide (SiC)
Accepted Aug 29, 2022 technologies and their cascode topology (SiC-Cascode), operating at high
switching frequencies. The analyzes are performed using a Boost converter
designed for conversion systems in the more-electric aircraft (MEA) context,
Keywords: where the alternating current (AC) power systems can operate at fixed
frequency (115 V/400 Hz), or at variable frequency (115 V/360-800 Hz),
DC distribution such as observe in Boeing 787, and the direct current (DC) power system
Microgrids can operate with a DC bus of 400 V and +/-270 V are normally used. To
More-electric aircraft validate the project, computer simulations were performed and a 1.0 kW
Power electronics prototype was built in the laboratory. The performance analyses
Silicon carbide demonstrates that 97.5% of efficiency is achieved at 500 kHz switching
Wideband gap semiconductors frequency.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Luiz Carlos Gomes Freitas
Faculty of Electrical Engineering, Federal University of Uberlandia
Av. João Naves de Ávila, 2121, Campus Santa Mônica – Bloco 3N, CEP 38400-902, Uberlândia, Brazil
Email: [email protected]

1. INTRODUCTION
The material traditionally applied for semiconductor purposes in power electronics converters is
silicon (Si), which was discovered at the end of the 19th century and has been dominant ever since, due to its
large consumer market and technology widely disseminated at industrial level [1]. Currently, several material
technologies besides Si are applicable in semiconductor devices. Some examples available on the market are
the silicon carbide (SiC) and gallium nitride (GaN), which are used in high-performance, wideband gap
(WBG) switching elements. The use of the above-mentioned materials, although initially costly, has
developed rapidly, seeking to reach the theoretical limits of each material.
Recent technologies of Si, SiC and GaN technologies have a higher energy band and breakdown
voltage enabling the creation of high voltage and WBG devices [2], [3]. Due to the high switching speed of
WBG devices, it is possible to increase the switching frequencies and to reduce the switching losses, leading
to greater energy savings. Additionally, it is possible to reduce the volume and weight of the passive
elements used, drastically reducing the amount of raw material needed to make the final product. The gain
with energy savings is considerable, offsetting the initial investment during its lifetime, even with an initial
higher cost when compared to the implementation of Si semiconductors [4], [5]. However, with the advances
in the semiconductor industry, this investment has been reduced enabling greater profitability in the
application of WBG semiconductors also in the short term [6]. Despite the advantages of WBG technologies,
the high switching frequencies, added to the high power processing levels, can accentuate the EMI [2], [7],

Journal homepage: http://ijpeds.iaescore.com

1952  ISSN: 2088-8694

[8]. Several previous studies reported in [4], [5], [9]–[15] prove the effective gain in performance when
WBG semiconductor devices are applied in power electronics converters, encouraging their increasing
application in the industrial and academic environment. Additionally, recent studies concerning the
deployment of SiC [16]–[21] and GaN [22], [23] technologies in several applications, the latter being used in
high-power devices, as reported in [24].
Currently, SiC technology moves to the preponderant position in high power applications since it
presents greater thermal conduction among the mentioned technologies, together with greater availability in
the market. This technology is already found in different devices, such as MOSFETs and JFETs, and in the
form of hybrids technologies, such as SiC-Cascode, originated from the use of combined SiC-JFET and Si-
MOSFET devices [25], which presents the advantages of simplified Si triggering, combined with the high
bandwidth of the SiC-JFET device. The technology begins to gain space in the market, being applied to high-
power modules by several manufacturers, in addition to being widely discussed in the academic context as a
superior alternative to Si. It has an intermediate price, with transistors accessible to the industry and to the
final consumer. Therefore, the evaluation of SiC-based semiconductor technologies in their practical limits is
encouraged, aiming at greater efficiency and higher power density from the increase in the switching
frequency. Such characteristics form the basis of requirements for embedded applications, such as found in
electric vehicles (EV) and more-electric aircraft (MEA) [26]–[29]. Applied to diodes, they are present in
devices with zero reverse recovery current and zero forward recovery voltage, which inhibits power losses
due to switching in passive devices [30].
Focusing on more-electric aircraft, in [27] the authors presented an important study evidencing that
the search for improvements in electrical converters is the key to systems with high efficiency and reliability.
Thus, the use of new semiconductor technologies (SiC and GaN) helps in the performance of these
converters, which are influenced by harsh environment, low pressures and cosmic radiation present at high
altitudes. In order to illustrate what has been exposed, Figure 1, which was originally presented in [27],
shows the main requirements and trends that must be analyzed in electronic converter designs for MEA
applications. Figure 2 indicates that the use of SiC-MOSFET and GaN-HEMT, with switching frequencies in
the range of 100 kHz to around 300 kHz, are the best choice.
Marroqui et al. [31] presented a comparative study between SiC MOSFET and SiC-Cascode in the
scope of robustness in short-circuit conditions. Both technologies were capable of with-standing 3 µs events
and showed little variation in their parameters after 200 occurrences of this type. Zhao et al. [32] presented
the semiconductor suitability for parallel operation has been evaluated and the selected SiC-MOSFET device
presented the highest efficiency in current division, from a double pulse test, at a continuous voltage of 400 V
and inductive load of 150 H. Martin et al. [33] their performance on the implementation of a high-power
inverter (100 kW), operating at a fixed switching frequency of 50 kHz has been analyzed. The SiC-Cascode
technology presented greater efficiency and provided higher power density. These studies have shown
comparisons between the technologies of semiconductors only on a restricted scale, emphasizing specific
operational characteristics and at predetermined switching frequencies.

~60 m

EXT-PWR 2 x115 V (AC)

90 kVA

Generator
~60 m 2 x 250 kVA
G G G G
ECSM Environmental Control Wing Ice Protection
System (ECS) System (WIPS)
ECSM M ECS
Load Generator
G
M M G Starter Generator Hydraulic Pump
M M
M G Emergency M
Inverter
M Generator (RAT) M Electro-Mechanical
M Actuator 235 V (AC)
EXT-PWR 2 x115 V (AC) Electro-Hydraulic
M ±270 V (DC)
90 kVA Actuator
Transformer Rectifier 115 V (AC)
28 V (DC)
115 V (AC) / 28 V (DC)
APU Generator
M
M
2 x 225 kVA
G G

M M M

Figure 1. Electric schematic of MEA–adapted from [27]

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
 Int J Pow Elec & Dri Syst ISSN: 2088-8694  1953
R f SE²A/M l

100 M Potential of Ga
N
Si an d
10 M lim SiC
it

Power (VA)
1M
100 k SiC-MOSFET
IGBT
10 k
1k GaN-HEMT
Si-MOSFET
100

100 1k 10 k 100 k 1 M 10 M 100 M

Frequency (Hz)

Figure 2. Semiconductors typical application–adapted from [27]

In this scenario, the article proposed herein presents a study concerning the performance analysis of
Boost converter operating as Front-end converter for Power Factor Correction (PFC) and also as DC-DC
converter for electronics loads, both in the MEA context, where the Alternating Current (AC) power systems
can operate at fixed frequency (115 V/400 Hz), or at variable frequency (115 V/360-800 Hz), such as in
Boeing 787, and the direct current (DC) power system can operate with a DC bus of 400 V and +/-270 V are
normally used. To validate the project, computer simulations were performed and a 1.0 kW prototype was
built in the laboratory. The performance analyses demonstrates that 97.5% of efficiency is achieved at
500 kHz switching frequency. To present the obtained results, the remainder of this article is structured as
follows. Section 2 describes the proposed power electronics converter for carrying out the simulation and
experimental tests, including its design procedure and control strategies. Section 3 details the methodology
used for efficiency analysis; and finally, the concluding remarks are elucidated in Section 4.

Fuse
Lb
Magnetic
2. IMPLEMENTATION OF A SIC CurrentAND
Sensing SIC-CASCODE BASED BOOST CONVERTER
594 uH

Mains Voltage In order to analyze the applicability of high frequency systems for highC1 power density power
electronics converters, which is essential for embedded systems such as MEA applications, 1000 uF
the performance
of the Boost converter is evaluated underDigitalswitching frequency varying from 50 kHz to 1300 kHz. The Resistive control
Load
Full Brigde Rectifier C2
algorithm was developed in C languageProcessor and embedded in the multi-core DSP TMS320F28379D from Texas
Signal
Isolated
Instruments. For DC-DC conversion system, the Boost Gate topology
Driver is portrayed in Figure
10003,
uF and the parameters

of the experimental set-up are described in Table 1. The system contains protection devices against voltage
surges (varistor) and overcurrent (fuse). Additionally, an overcurrent protection was implemented via
software and is based on tripping aid sent by the current sensor, assuring safety and reliability under short-
Voltage
circuit occurrence. For the AC-DC conversion Sensing system, a Graetz bridge rectifier is added at the input of the
circuit, with the other parameters unchanged. The average current mode control technique was implemented
via software in the DSP. The AC input voltage is 220 V RMS at frequencies varying from 400 Hz to 800 Hz, as
found in MEA’s electric power systems (EPS).

Fuse
Lb
Magnetic
Current Sensing
594 uH
C1
DC Source

1000 uF
Resistive
Digital Load
C2
Signal
Processor Isolated
Gate Driver 1000 uF

Voltage
Sensing

Figure 3. Boost converter for AC-DC operation (with a Graetz bridge rectifier added at the input) and for
DC-DC operation (without a Graetz bridge)
Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1954  ISSN: 2088-8694

Considering the application in high voltage and high switching frequencies, optimization criteria
based on reference designs made available by semiconductor manufacturing companies (Cree, Texas
Instruments, United SiC, among others) were implemented for the PCB. An isolated DC-DC converter for
the gate driver was also added to the circuit board, reinforcing the galvanic isolation between the control
circuits and the power circuit. Other components were powered by dedicated integrated circuits implemented
in the PCB. The final board kept dimensions of 100×100 mm, with semiconductor devices kept at the bottom
of the board providing better connection to the heatsink.

Table 1. Parameters of the Boost converters implemented in laboratory

Parameter DC-DC conversion AC-DC conversion
Rated power [W] 1000 1000
Input voltage 200 𝑉𝐷𝐶 220 𝑉𝑅𝑀𝑆 - 400 Hz to 800 Hz
Output voltage [V] 400 400
Capacitor CB 1 [μF] 500 500
Inductor Lb [μH] 594 594
R ESR(C)1 [mW] 300 300
Resistive load [W] 160 160
Diode C3D10065A C3D10065A
C3M0120100K C3M0120100K
Switches
UF3C120150K4S UF3C120150K4S
Switching frequency 200 kHz 200 kHz

Inductor physical specification

Core material TDK SIFERRIT N97 TDK SIFERRIT N97
Core PQ 50/50 PQ 50/50
Gap [mm] ~1 ~1
Gap Material Paper Paper
R ESR(L) [mW] 37 37
Turns 45 45
Wire AWG 18 x 2 AWG 18×2

Gate driver parameters

R G(on) [W] 10 11.7
R G(off) [W] 3.2 23.5
CGS(ext) [pF] 1000 1000
1
Values related to the equivalent total capacitance

In the literature, numerous control strategies applicable to power electronics converters are found,
including from simplified controls based on hysteresis or pulse width modulation (PWM) [34], [35], to
asymmetric PWM controllers [36] and the use of multi-objective optimization algorithms [37]. Among the
strategies applicable to AC-DC conversion systems, there is the average current mode control technique [38],
which was implemented in this work to control voltage and current simultaneously, in order to impose input
line current with low total harmonic distortion (THDi), which is a crucial indicator in electrical power quality
indicators in power electronics systems [39]. A sinusoidal voltage reference signal synchronized with the
input voltage is also needed to assure unitary power factor. This is made deploying a second order
generalized integrator - phase locked loop (SOGI-PLL) [40], which filters potential noise and input voltage
distortions. The system bandwidth must be low, so as not to interfere with the waveform imposed by SOGI-
PLL at the input current. A first-order low-pass filter was also deployed in the voltage feedback to attenuate
the current ripple caused by DC-bus voltage ripple at the input voltage frequency.
For the design of the controllers, the converter transfer function is required for small disturbances in
the duty cycle. From the average model in the state-space, the plant was raised, described in (1) for the output
voltage, and in (2) for the input current. The parameters used referred to the design values, with an average
duty cycle of 0.5 and load resistance of 160 .

𝑣̂(𝑠)
𝑜 −1×104 𝑠 3 +6.721×108 𝑠 2 +9.537×10−5 𝑠+5.649×1014
𝐺1 (𝑠) = = (1)
𝑑̂ (𝑠) 𝑠 4 +25𝑠 3 +1.681×106 𝑠 2 +2.101×107 𝑠+7.062×1011

𝑖𝑛 (𝑠)
𝑖̂ 6.723×105𝑠 3 +2.521×107 𝑠 2 +5.651×1011 𝑠+1.412×1013
𝐺2 (𝑠) = = (2)
𝑑̂ (𝑠) 𝑠 4 +25×𝑠 3 +1.681×106 𝑠 2 +2.101×107 𝑠+7.062×1011

Using MATLAB software, the PI controllers were designed from the closed loop frequency
response. Each controller has its estimated bandwidth according to the switching frequency and the input
voltage frequency (estimated bandwidth greater than 10–100 times the input voltage frequency and less than

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1955

1/10 of the switching frequency, approximately, for the current controller; and at least 10 times lower
bandwidth for the voltage controller).
In (3) and (4) are found the transfer functions of the applied PI controllers, for the control of current
and voltage, respectively, in the frequency domain. Controller C1 refers to the current loop compensation,
found in the internal control loop, while controller C2 refers to the external PI controller, after voltage
feedback, both shown in Figure 4.
𝑠+7.507×104
𝐶1 (𝑠) = 0.16745 ( ) (3)
𝑠

𝑠+44.42
𝐶2 (𝑠) = 0.094203 ( ) (4)
𝑠

For the application in the discrete-time domain, from the DSP, the controllers 𝐶1 and 𝐶2 if were
emulated, using the Tustin method. The method maintains its transient characteristics as previously designed
[41], only varying the sampling frequency in the emulation process, which depends on the switching
frequency of the converter. The block diagram illustrating the implemented control strategy in Figure 3 and
the power circuit shown in Figure 4 were validated using PSIM software. In Figure 5, for a load step from
0.5 pu to 1 pu (320  to 160  of load), a voltage undershoots of approximately 5% and an accommodation
time (1%) of about 70 ms were measured, with imposition of sinusoidal current with low harmonic distortion
(THDi < 5% for rated load), validating the designed control. As an example, in Figure 5 it is noted that the
DC voltage at the output of the Boost converter suffers little variation due to the action of the voltage and
current controllers. For operation at 400 Hz and 800 Hz, the bandwidth of the PI controllers does not satisfy
the operating conditions for switching frequencies below 100 kHz. The respective experimental results will
be presented in section 4.

SOGI
PLL[z] ADC K1 VIN(t)
Voltage
Sensor

|x|
|x| K4 ADC K2 IIN(t)
Digital Current
Gain Sensor

Output
× - C1[z] Gate G(s) Voltage
VREF +- C2[z] + Driver
Boost
Converter

LPF[z] K5 ADC K3
Digital Voltage
Gain Sensor
Digital Control Analog Control

Figure 4. Experimental and simulation control setup

Output Voltage (V)
Input Current (A)

Time (s)

Figure 5. Simulation result for a load step at 400 Hz from 50% to 100% of rated power
Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1956  ISSN: 2088-8694

3. LOSSES ANALYSIS FOR THE IMPLEMETED BOOST CONVERTERS

Regarding the losses in power electronics converters, it is possible to highlight the conduction
losses, switching losses and magnetic losses. Leakage losses, although they exist, do not contribute
significantly to the studies carried out, so they will be disregarded. The device’s parameters used are usually
found in their datasheet, and in some cases, they must be measured from laboratory tests, as in the case of the
inductor series resistance R ESR(L) . The conduction power losses in switched circuits can be estimated from
the resistance of the elements used, as well as from the current circulating in the circuit to be analyzed, with
the exception of diodes, which also depend on their forward bias voltage. The other power conduction losses,
such as in the PCB tracks or in connection cables, have a reduced contribution (complementary losses) and will
not be addressed in the proposed analysis. Therefore, the conduction power losses in a FET device can be
expressed as (5).

𝑃𝑐𝑜𝑛𝑑(𝑆) (𝑡) = 𝑅𝐷𝑆(𝑜𝑛) (𝑇𝐽 )𝑖𝑆2 (𝑡) (5)

here, 𝑖𝑆 is the instantaneous drain current, 𝑅𝐷𝑆(𝑜𝑛) is the resistance between drain and source in the on state
and 𝑇𝑗 is the junction temperature of the transistor.
Switching losses occur during semiconductor state transitions, being characterized by the superposition
between voltage and current at this instant. As the number of transitions is directly proportional to the switching
frequency, switching losses also have the same characteristic. However, switching losses have greater modeling
complexity, requiring the contribution of several studies in order to develop a model that is close to the one
found in practice. Prado et al. [42] presented an analytical model for MOSFETs suitable for this application,
therefore, it was used in the proposed paper. Therefore, losses for MOSFETs can be given by (6).
1
𝑃𝑠𝑤(𝑆) = (𝑡𝑜𝑛 𝐼𝑜𝑛 + 𝑡𝑜𝑓𝑓 𝐼𝑜𝑓𝑓 )𝑓𝑐ℎ 𝑉𝐷𝑆 (6)
2

where 𝑃𝑠𝑤(𝑆) represents the switching power dissipation in the MOSFET, 𝑡𝑜𝑛 and 𝑡𝑜𝑓𝑓 are the overlap times
in the transitions for conduction and blocking respectively, 𝐼𝑜𝑛 and 𝐼𝑜𝑓𝑓 are the input and output switching
currents, 𝑓𝑐ℎ is the switching frequency and 𝑉𝐷𝑆 is the blocking voltage at the switch.
To determine the overlap times, is necessary to determine the charging and discharging behavior of
the gate-source capacitance (𝐶𝐺𝑆 ) and gate-drain capacitance (𝐶𝐺𝐷 ). The modeling of the gate-source load
(𝑄𝐺𝑆 ) is given by (7) and is based on the input capacitance (𝐶𝐼𝑆𝑆 ) of the MOSFET, the gate threshold voltage
(𝑉𝑇𝐻 ) and the gate plateau voltage (𝑉𝑃𝐿 ).

𝑄𝐺𝑆 = 𝐶𝐼𝑆𝑆 (𝑉𝑃𝐿 − 𝑉𝑇𝐻 ) (7)

For the gate-drain load model (𝑄𝐺𝐷 ), the study becomes complex due to the non-linearities of 𝐶𝐺𝐷
with the variation of the blocking voltage. From [42] one can conclude that a valid gate-drain load model
(𝑄𝐺𝐷 ) can be found using (8), and takes into account only two points of the capacitance curve 𝐶𝐺𝐷 x𝑉𝐷𝑆 found
in the datasheet. The determination of point A was associated with the time constant of RC circuits (τ)
representing the voltage drop of 𝑉𝐷𝑆 in 2τ, where the voltage reaches 13.5 % of 𝑉𝐷𝑆 . The proposed method is
based on a linearization between the voltages of points A and B (0.135× 𝑉𝐷𝑆 and 𝑉𝐷𝑆 , respectively), resulting
in the approximation by the area of a triangle, equivalent to the capacitance charge 𝐶𝐺𝐷 .
𝐶𝐺𝐷(𝐵) 𝑉𝐷𝑆 +𝐶𝐺𝐷(𝐴) 0.135𝑉𝐷𝑆
𝑄𝐺𝐷 = (8)
2

𝑄 = 𝑄𝐺𝑆 + 𝑄𝐺𝐷 (9)

Knowing the two contributions to the total gate-source load, the total load (Q) required to activate the switch
is reached according to (9).
For the calculation of the overlap times, the gate currents for charging (𝐼𝐺𝑜𝑛 ) and discharging
(𝐼𝐺𝑜𝑓𝑓 ), still necessary, which require the internal gate resistors (𝑅𝐺𝑖𝑛𝑡 ), for switching on (𝑅𝐺𝑜𝑛 ) and for
switching off (𝑅𝐺𝑜𝑓𝑓 ), as given by (10) and (11).
𝑉𝐷𝑅 −𝑉𝑃𝐿
𝐼𝐺𝑜𝑛 = (10)
𝑅𝐺𝑜𝑛 +𝑅𝐺𝑖𝑛𝑡

𝑉𝑃𝐿
𝐼𝐺𝑜𝑓𝑓 = (11)
𝑅𝐺𝑜𝑓𝑓 +𝑅𝐺𝑖𝑛𝑡

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1957

Finally, the overlap times 𝑡𝑜𝑛 e 𝑡𝑜𝑓𝑓 can be calculated from (12) and (13), which are used in (6) to
determine the switching losses.
𝑄
𝑡𝑜𝑛 = (12)
𝐼𝐺𝑜𝑛

𝑄
𝑡𝑜𝑓𝑓 = (13)
𝐼𝐺𝑜𝑓𝑓

The switching losses referred to the diode are related to reverse recovery. Therefore, due to the use
of SiC technology devices, which present zero reverse recovery charge, it will be disregarded in this paper.
For the inductor, the core losses are dependent on the variation of magnetic flux, temperature, and the
switching frequency imposed on the device. According to the material used, curves that determine the
volume losses (𝑃𝑣 ) of the material can be presented by the manufacturers, according to the indicated
parameters. Losses are generally given in kW/m³, and are presented for discrete values of magnetic flux
variation. According to [43], the variation of magnetic flux for inductors (Δ𝐵) is given by (14), where N is
the number of turns, Δ𝐼𝐿 is the current ripple in the inductor, 𝑙𝑔 is the gap length, MPL is the mean magnetic
path length and 𝜇𝑚 is the relative magnetic permeability of the material.
Δ𝐼
0.4𝜋𝑁 2𝐿10−2
Δ𝐵 = 𝑀𝑃𝐿 (14)
𝑙𝑔 +( )
𝜇𝑚

𝑃𝑐𝑜𝑟𝑒 = 𝑃𝑣 𝑉𝑂𝐿𝐿 (15)

From the data collected, the inductor loss is calculated using (15), where 𝑃𝑐𝑜𝑟𝑒 is the power dissipated in the
magnetic core, and 𝑉𝑂𝐿𝐿 is the inductor volume given in m³.
For DC-DC conversion, the analysis of power losses will be performed using switching frequencies
between 100 kHz and 500 kHz, corresponding to the typical and borderline frequency for SiC technology,
seen in Figure 2. The analysis will be carried out in order to consider the sources of losses and indicate
possible points for improving the overall efficiency, making it possible to achieve greater efficiency and
power density for embedded applications, as in the MEA context. Operating temperature of 100°C is
considered for the determination of power losses in the MOSFETs (SiC and SiC-Cascode), and 40°C is
considered for the rest of the circuit. Using the mathematical procedure described in the last section it is
possible to quantify the power losses and also identify the mains sources of power in power electronics
converters.
As portrayed in Figure 6, for a switching frequency of 100 kHz, the current ripple in the inductor is
higher, causing significant losses in the magnetic core, which is equivalent to 30% of the total losses. The
switching power losses at this frequency correspond to 16% (13.62 W). For 500 kHz switching frequency,
the core losses are drastically reduced, as a consequence of the lower current ripple and magnetic flux in the
inductor. Hence, they are responsible for only 1% of the total losses. In contrast, switching losses are
dominant, equivalent to 61%. The conduction losses found have little variation, since the analysis is
performed under the same load condition.
For SiC-Cascode, the results obtained are illustrated in Figure 7. The main differences are found in
the semiconductors where the power switching loss and the power conduction loss are lower using
SiC-Cascode device. For 100 kHz switching frequency, the power switching losses are equivalent to only
10% of the total power loss; while at 500 kHz, the total power loss is equivalent to 42%, with linear growth
as the switching frequency increases. Even though the same inductor is used in both experiments, one must
observe that the losses in the magnetic material were reduced from 30% to only 2% of the total. This is due to
the fact that the current ripple is considerably reduced in higher switching frequencies. The total power loss
at switching frequencies of 100 kHz and 500 kHz are 13.87 W and 14.84 W respectively. The reduction in
the power switching losses is due to the simplification of the transistor's drive characteristics, which has
higher VTH and lower VPL, leading to a shorter current transition period and waveforms overlap. The V PL
voltages measured for SiC device was 9.5 V and 6.52 V for SiC-Cascode device. In this context, it is
important to emphasize that this operational characteristic together with the intrinsic capacitances, dictate the
switching speeds, directly affecting the bandwidth of the transistor. Therefore, special attention must be
dedicated when choosing FETs for designs with high switching frequency.

Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1958  ISSN: 2088-8694

Psw(M)
Pcore
100 kHz 500 kHz Pcond(S)
6.86% 4.89% Pcond(D)
15.79%
9.94% Pcond(C)
14.55% Pcond(L)

14.54%

30.39% 60.50%
20.19% 8.73%
1.39%
12.23%
Total = 13 .62 W Total = 18. 91 W
η = 98 . 66% η = 98 .14%

Figure 6. Power losses analysis with SiC device for rated power (1 kW)

Psw(M)
Pcore
100 kHz 500 kHz Pcond(S)
9.53% 6.73% 6.24% Pcond(D)

12.66% Pcond(C)
14.28% Pcond(L)

42.47%
29.84%
18.53%
19.83%

1.77% 18.33%
19.79%
Total = 13.87 W Total = 14 .84 W
η = 98 .63% η = 98 .53%

Figure 7. Power losses analysis with SiC-cascode device for rated power (1 kW)

4. EXPERIMENTAL RESULTS AND DISCUSSIONS

Figure 8 shows the laboratory prototype. The connections between the control and power circuits
were made from shielded cables to mitigate electromagnetic interference. The connection of external
components, such as voltage sensors, protection devices, inductor and bridge rectifier were made from
ordinary cables. For heat dissipation, an aluminum heatsink is used at the bottom of the power circuit. The
converter performance was tested in two operating situations: DC-DC conversion and AC-DC conversion at
varying frequency of 400-800 Hz, as found in MEA. The results were obtained in steady state, with stable
operating temperature according to the load and switching frequency. For all operating conditions, the input
voltage was applied by a controlled voltage source in order to mitigate injected noise and maintain harmonic
distortion levels without interference from external factors. The efficiency measurements refer to the input
and output powers of the power circuit, disregarding losses related to gate-driver and control circuits. In this
section, the results obtained for both operating conditions will be discussed, seeking to highlight the main
advantages provided by the semiconductors technologies used and their potential for MEA application.

Input Inductor
Graetz Bridge Rectifier
Output Capacitors

DSP / Voltage
Sensing
Power Board

Figure 8. Prototype developed in laboratory

4.1. DC-DC conversion system for MEA

For the DC-DC conversion system, the input and output powers of the experimental were measured
using a precision equipment - YOKOGAWA WT230. The switching frequency was set from 50 kHz to

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1959

1300 kHz. The maximum switching frequency was chosen regarding the maximum power dissipation of 50
W, which is reached at switching frequency of 1300 kHz, at rated power. The waveforms presented were
measured using the TEKTRONIX MDO3034 and MDO4054-3 oscilloscopes, using isolated voltage probes,
in order to mitigate external noise. For reasons of safety and susceptibility to mutual interference between the
control and power systems, in conjunction with operation at the practical limits of the semiconductors, the
DC-DC operation will be driven in open loop. However, small adjustments are made according to the
operating frequency in order to keep the gain constant. From the data obtained, it is observed that the SiC-
Cascode MOSFET technology provided greater efficiency with switching frequency of 100 kHz, as
illustrated in Figure 9. Initially, both technologies proved to be efficient, reaching 98.8% efficiency at rated
load. The higher the switching frequency, the greater the discrepancy between the performance of Sic and
SiC-Cascode devices. This caused due to switching losses which is more significant using SiC devices.
During operation at higher switching frequency, the rise and fall times of the V DS voltage were
measured, as seen in Figure 10. As expected, a difference between the switching times was verified, with the
SiC-cascode device being faster in the transitions. Due to this feature, the overlap time between current and
voltage is reduced, causing less switching power loss, and achieving greater overall efficiency. This occurs
due to the simplified characteristics of the activation of the SiC-Cascode switch, with reduced output
capacitance (Coss) of the transistor at rated DC-bus voltage (400 V), being 32 pF against 52 pF of the SiC
switch, approximately. Therefore, for both technologies, reduced switching times were observed, which lead
to lower power switching loss. This characteristic highlights the importance of the article presented herein for
the advancement of MEA since it demonstrates that high switching frequencies should be chosen for the
design of DC-DC conversion systems with reduced weight, leading to the reduction of fuel consumption in
the aircraft.

Figure 9. Efficiency x Frequency, DC-DC conversion system

9.2 ns 9.9 ns

(a) (b)

8.96 ns
9.3 ns

(c) (d)

Figure 10. Experimental results; switching time for SiC MOSFET: (a) rise time and (b) fall time: switching
time for SiC-Cascode MOSFET, (c) rise time, and (d) fall time

Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1960  ISSN: 2088-8694

4.2. AC-DC conversion system for MEA

Figure 11 shows the main waveforms observed in the laboratory with AC power supply frequency
of 400 Hz and 800 Hz and switching frequency of 200 kHz. The waveforms were acquired from the
TEKTRONIX MDO3034 and MDO4054-3 oscilloscopes, together with the TEKTRONIX TCPA-300
current sensor. The data for measuring harmonic distortion were obtained from oscilloscopes, with a
sampling of 1 Msamples/s, stored in an external device and processed computationally via MATLAB. Up to
the 41st harmonic orders were considered, and data equivalent to three fundamental periods sampled in each
case were treated.
For both AC power supply frequencies, low harmonic distortion of current and high-power factor is
achieved. From the analyses of the waveforms, it is also possible to observe that there is more significant
distortion in the voltage and current waveforms when SiC-Cascode device is used. This arise from the high
dv/dt indexes inserted by the fast-switching technology, but can be mitigated by raising the drive resistors,
with a consequent reduction in the switching speed and efficiency [2]. This feature sets up a conflict of
choice between distortion minimization and circuit efficiency, and it is up to the designer to find the optimal
operating point, which changes according to the technology deployed. This difference, although subtle, can
lead to problems with EMI, and should be explored through practical analysis in future works.

vin vin
Vout Vout

iin iin

(a) (b)

vin vin

Vout Vout

iin iin

(c) (d)

Figure 11. Waveforms, MEA AC-DC operation with 400 Hz: (a) SiC MOSFET, (b) SiC-Cascode MOSFET,
(c) SiC MOSFET, and (d) SiC-Cascode MOSFET

For the efficiency analysis, the results presented herein were obtained varying the switching
frequency between 100 kHz and 1000 kHz. Figure 12 demonstrates that better results are achieved with the
SiC-Cascode device. Peak efficiency for AC power supply frequency of 400 Hz was achieved at 100 kHz
switching frequency, reaching 98.12%, against 98.00% obtained with SiC device. For the higher analyzed
switching frequency (1000 kHz), the SiC-Cascode technology proved to be superior, with 96.2% efficiency
against 94.93% achieved with the SiC technology. Such a reduction in efficiency, compared to DC-DC
conversion system, comes from the insertion of the diode rectifier bridge at the input of the circuit, which
causes additional power losses. For operation with AC power supply frequency at 800 Hz, the results are
similar.
Figure 13 presents the main obtained results for a wide load range and with 400 Hz AC power
supply frequency and 250 kHz switching frequency. With SiC-Cascode device the achieved efficiency level
is higher when compared to results obtained with SiC device. For the power factor, measured from
YOKOGAWA WT230 precision equipment, the results obtained remained above 0.96 for both technologies,

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1961

increasing according to the load level and reaching a peak value above 0.99 at rated load. Due to faster
switching conditions, SiC-Cascode technology introduces more noise into the system as observed in
Figure 11, leading to a slight variation in the achieved power factor. Consequently, a degradation in the
harmonic distortion index of the input current can be observed for this technology, with about 16% of
distortion for the lowest load condition (22% of the rated power), decreasing to 5% at rated power. In
contrast, SiC technology did not present such problems, providing a THDi of 7% for the lowest load
condition and 2% at rated power.

(a) (b)

Figure 12. Efficiency x frequency, AC-DC operation with different AC power supply frequency:
(a) 400 Hz and (b) 800 Hz

(a)

(b) (c)

Figure 13. Experimental results for wide load range with 400 Hz AC power supply frequency: (a) efficiency,
(b) power factor, and (c) total harmonic distortion of the input line current

With 800 Hz AC power supply frequency, the results obtained at rated load confirm a THDi of
about 3% (SiC) and 5% (SiC-Cascode), and power factor above 0.99. As the 800 Hz AC power supply
frequency is used sporadically in aircraft systems, the results for different load conditions will be analyzed
only at nominal AC power supply frequency (400 Hz). The dynamic response of the output voltage controller
and the Boost input current controller, when subjected to a 50% load step down and over, can be seen from
Figure 14. They resulted in settling times of about 60 ms under both applied conditions, and percentage
overshoot/undershoot of approximately 7% during dynamic tests.

Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1962  ISSN: 2088-8694

Vout Vout
iin vin vin
iin
(a) (b)

Figure 14. Dynamic response at 400 Hz (200 V/div, 5 A/div, and 20 ms/div):
(a) load step from 50% to 100% of rated power and (b) load step from 50% to 100% of rated power

As a specific condition found in MEA systems, the dynamic variation of the input voltage frequency
was also addressed. Applying an increasing voltage frequency ramp from 400 Hz to 800 Hz, during 200 ms,
it is possible to observe that the SOGI-PLL operation accurate. The system was able to follow the frequency
variation, keeping the imposition of the input current synchronized with the ac voltage. The output voltage
remained regulated at 400 V throughout the entire transient period, validating the voltage regulation for the
frequency ramp. Figure 15 shows the behavior of the structure facing variation of the grid power supply
frequency, proving that a perfect synchronism between current and voltage is achieved. As observed, the
system was able to maintain synchronism under frequency variation, maintaining power factor correction and
load voltage regulation.

Vout iin

vin

Figure 15. Synchronism between grid voltage and current during frequency variation (from the left to the
right: from 400 Hz to 800 Hz, during 200 ms)

5. CONCLUSIONS
The mathematical description of power losses in power electronics converters deploying SiC and
SiC-Cascode devices is described in this paper, and it was corroborated by experimental analysis. It has been
demonstrated that Wide-bandgap semiconductors can achieve high switching speed and high efficiency even
at switching frequencies above 1.0 MHz. Both technologies provide reduced switching times, which lead to
lower power switching loss. This characteristic highlights the importance of the article presented herein for

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1963

the advancement of More Electric Aircraft, since it demonstrates that high switching frequencies should be
chosen for the design of DC-DC and AC-DC conversion systems with reduced weight, leading to the
reduction of fuel consumption in the aircraft.
For DC-DC conversion systems, the SiC-Cascode MOSFET has simplified drive characteristics,
with lower gate-driver requirement, since lower level of peak current is needed. With SiC-Cascode device,
the main differences are found in the power switching loss and the power conduction loss. For 100 kHz
switching frequency, the power switching losses are equivalent to only 10% of the total power loss; while at
500 kHz, the total power loss is equivalent to 42%, with linear growth as the switching frequency increases.
The reduction in the power switching losses is due to the simplification of the transistor's drive
characteristics, which has higher VTH and lower VPL, leading to a shorter current transition period and
waveforms overlap. In this context, it is important to emphasize that this operational characteristic together
with the intrinsic capacitance values, dictate the switching speeds, directly affecting the bandwidth of the
transistor. Therefore, special attention must be dedicated when choosing FETs for designs with high
switching frequency.
On the other hand, for AC-DC conversion systems, SiC-Cascode technology introduces more noise
into the system, leading to a significant degradation in the harmonic distortion index of the input current.
This arise from the high dv/dt indexes inserted by the fast-switching technology, but can be mitigated by
raising the drive resistors, with a consequent reduction in the switching speed and efficiency. This feature
sets up a conflict of choice between distortion minimization and circuit efficiency, and it is up to the designer
to find the optimal operating point, which changes according to the technology deployed. The difference in
THDi, although subtle, can lead to problems with EMI, and should be explored through practical analysis in
future work. In this context, the THDi achieved was above 5% for a wide load range. For the lowest load
condition (22% of the rated power), a THDi of 16% was observed, decreasing to 5% at rated power. In
contrast, SiC technology did not present such problems, providing a THDi of 7% for the lowest load
condition and 2% at rated power. Therefore, since SiC technology presents greater stability in its parameters,
lower EMI irradiation can be assured using this device. In conclusion, one can observe that SiC and SiC-
Cascode devices can provide high efficiency and high-power density to energy conversion systems found in
MEA’s Electric Power Systems. However, the commutation speed of semiconductors must be carefully
analyzed in order to reach the optimal relation between EMI and efficiency.

ACKNOWLEDGEMENTS
This research was partially financed by National Council for Scientific and Technological
Development (CNPq) under grants 303350/2019-9 and 147857/2021-0; the Coordination for the
Improvement of Higher Education Personnel (CAPES); the Research Support Foundation of the State of
Minas Gerais (FAPEMIG) under grant PPM-00485-17, and the Federal University of Uberlandia (UFU),
under process 23117.060675/2022-10. The authors would like to thank Texas Instruments for the ICs
samples, and Professors João B. Vieira Junior and Luiz C. Freitas for supporting this work through grants
304489/2017-4 and 304479/2017-9 of the National Council for Scientific and Technological Development
(CNPq).

REFERENCES
[1] V. Benda, N. Barsoum, J. F. Webb, and P. Vasant, “Power semiconductors—state of art and future trends,” Sarawak, (Malaysia),
2011, pp. 16–24. doi: 10.1063/1.3592437.
[2] X. Yuan, I. Laird, and S. Walder, “Opportunities, Challenges, and Potential Solutions in the Application of Fast-Switching SiC
Power Devices and Converters,” IEEE Trans. Power Electron., vol. 36, no. 4, pp. 3925–3945, Apr. 2021, doi:
10.1109/TPEL.2020.3024862.
[3] J. Millan, P. Godignon, X. Perpina, A. Perez-Tomas, and J. Rebollo, “A Survey of Wide Bandgap Power Semiconductor
Devices,” IEEE Trans. Power Electron., vol. 29, no. 5, pp. 2155–2163, May 2014, doi: 10.1109/TPEL.2013.2268900.
[4] R. Ø. Nielsen, L. Török, S. Munk-Nielsen, and F. Blaabjerg, “Efficiency and cost comparison of Si IGBT and SiC JFET isolated
DC/DC converters,” in IECON 2013 - 39th Annual Conference of the IEEE Industrial Electronics Society, Nov. 2013, pp. 695–
699. doi: 10.1109/IECON.2013.6699219.
[5] M. G. H. Aghdam and T. Thiringer, “Comparison of SiC and Si power semiconductor devices to be used in 2.5 kW DC/DC
converter,” in 2009 International Conference on Power Electronics and Drive Systems (PEDS), Nov. 2009, pp. 1035–1040. doi:
10.1109/PEDS.2009.5385745.
[6] M. Telford, “SiC’s power cuts cost,” III-Vs Review, vol. 16, no. 8, pp. 44–47, Nov. 2003, doi: 10.1016/S0961-1290(03)01182-7.
[7] B. Zhang and S. Wang, “A survey of EMI research in power electronics systems with wide-bandgap semiconductor devices,”
IEEE Journal of Emerging and Selected Topics in Power Electronics, vol. 8, no. 1, pp. 626–643, Mar. 2020, doi:
10.1109/JESTPE.2019.2953730.
[8] P. Bogónez-Franco and J. B. Sendra, “EMI comparison between Si and SiC technology in a boost converter,” in International
Symposium on Electromagnetic Compatibility - EMC EUROPE, Sep. 2012, pp. 1–4. doi: 10.1109/EMCEurope.2012.6396739.
[9] A. M. S. Al-bayati, S. S. Alharbi, S. S. Alharbi, and M. Matin, “A comparative design and performance study of a non-isolated
DC-DC buck converter based on Si-MOSFET/Si-Diode, SiC-JFET/SiC-schottky diode, and GaN-transistor/SiC-Schottky diode

Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)
1964  ISSN: 2088-8694

power devices,” in 2017 North American Power Symposium (NAPS), Morgantown, WV, Sep. 2017, pp. 1–6. doi:
10.1109/NAPS.2017.8107192.
[10] B. Rubino, G. Catalisano, L. Abbatelli, and S. Buonomo, “Comparative analysis of driving approach and performance of 1.2 kV
SiC MOSFETs, Si IGBTs, and normally-off SiC JFETs.” STMicroelectronics, 2015. Accessed: Apr. 27, 2022. [Online].
Available: https://www.st.com/resource/en/technical_article/ta0349-comparative-analysis-of-driving-approach-and-performance-
of-12-kv-sic-mosfets-si-igbts-and-normallyoff-sic-jfets-stmicroelectronics.pdf
[11] R. M. Burkart and J. W. Kolar, “Comparative evaluation of SiC and Si PV inverter systems based on power density and efficiency
as indicators of initial cost and operating revenue,” in 2013 IEEE 14th Workshop on Control and Modeling for Power Electronics
(COMPEL), Jun. 2013, pp. 1–6. doi: 10.1109/COMPEL.2013.6626462.
[12] M. M. Swamy, J.-K. Kang, and K. Shirabe, “Power loss, system efficiency, and leakage current comparison between Si IGBT
VFD and SiC FET VFD with various filtering options,” IEEE Trans. on Ind. Applicat., vol. 51, no. 5, pp. 3858–3866, Sep. 2015,
doi: 10.1109/TIA.2015.2420616.
[13] O. M. Rodríguez-Benítez et al., “Comparative performance and assessment study of a current-fed DC-DC resonant converter
combining Si, SiC, and GaN-based power semiconductor devices,” Electronics, vol. 9, no. 11, p. 1982, Nov. 2020, doi:
10.3390/electronics9111982.
[14] F. M. Shah, S. Maqsood, R. Damaševičius, and T. Blažauskas, “Disturbance rejection and control design of MVDC converter
with evaluation of power loss and efficiency comparison of SiC and Si based power devices,” Electronics, vol. 9, no. 11, p. 1878,
Nov. 2020, doi: 10.3390/electronics9111878.
[15] M. F. Yaakub, M. A. M. Radzi, F. H. M. Noh, and M. Azri, “Silicon carbide power device characteristics, applications and
challenges: an overview,” IJPEDS, vol. 11, no. 4, p. 2194, Dec. 2020, doi: 10.11591/ijpeds.v11.i4.pp2194-2202.
[16] S. Bhattacharya, C. Willich, and J. Kallo, “Design and Demonstration of a 540 V/28 V SiC-Based Resonant DC–DC Converter
for Auxiliary Power Supply in More Electric Aircraft,” Electronics, vol. 11, no. 9, p. 1382, Apr. 2022, doi:
10.3390/electronics11091382.
[17] W. Alves, L. Morais, and P. Cortizo, “Design of an highly efficient AC-DC-AC three-phase converter using SiC for UPS
applications,” Electronics, vol. 7, no. 12, p. 425, Dec. 2018, doi: 10.3390/electronics7120425.
[18] B. Hanko, M. Frivaldsky, and J. Morgos, “Evaluation of the efficiency performance of 3-phase, 6-switch PFC circuit based on the
used 1.2 kV SiC transistor,” Electronics, vol. 11, no. 3, p. 363, Jan. 2022, doi: 10.3390/electronics11030363.
[19] M. R. Rogina, A. Rodríguez, A. Vázquez, D. G. Lamar, and M. M. Hernando, “Event-focused digital control to keep high
efficiency in a wide power range in a SiC-based synchronous DC/DC boost converter,” Electronics, vol. 9, no. 12, p. 2154, Dec.
2020, doi: 10.3390/electronics9122154.
[20] J. Rąbkowski, A. Łasica, M. Zdanowski, G. Wrona, and J. Starzyński, “Portable DC supply based on SiC power devices for high-
voltage marx generator,” Electronics, vol. 10, no. 3, p. 313, Jan. 2021, doi: 10.3390/electronics10030313.
[21] R. Kopacz, M. Harasimczuk, B. Lasek, R. Miśkiewicz, and J. Rąbkowski, “All-SiC ANPC submodule for an advanced 1.5 kV EV
charging system under various modulation methods,” Energies, vol. 14, no. 17, p. 5580, Sep. 2021, doi: 10.3390/en14175580.
[22] C.-T. Ma and Z.-H. Gu, “Review of GaN HEMT applications in power converters over 500 W,” Electronics, vol. 8, no. 12, p.
1401, Nov. 2019, doi: 10.3390/electronics8121401.
[23] M. J. Carra, H. Tacc, and J. Lipovetzky, “Performance evaluation of GaN and Si based driver circuits for a SiC MOSFET power
switch,” IJPEDS, vol. 12, no. 3, p. 1293, Sep. 2021, doi: 10.11591/ijpeds.v12.i3.pp1293-1303.
[24] “GAN-HIGHPOWER,” GaN-Highpower | Hochschule Bonn-Rhein-Sieg (H-BRS), 2021. https://www.h-brs.de/de/gan-highpower
(accessed May 01, 2022).
[25] J. Bendel, “Cascode configuration eases challenges of applying SiC JFETs.” United Silicon Carbide, Inc., Mar. 2016. Accessed:
Apr. 27, 2022. [Online]. Available: https://unitedsic.com/appnotes/USCi_AN0004-Cascode-Configuration-Eases-Challenges-of-
Applying-SiC-JFETs.pdf
[26] K. Ni et al., “Electrical and Electronic Technologies in More-Electric Aircraft: A Review,” IEEE Access, vol. 7, pp. 76145–
76166, 2019, doi: 10.1109/ACCESS.2019.2921622.
[27] H. Schefer, L. Fauth, T. H. Kopp, R. Mallwitz, J. Friebe, and M. Kurrat, “Discussion on electric power supply systems for all
electric aircraft,” IEEE Access, vol. 8, pp. 84188–84216, 2020, doi: 10.1109/ACCESS.2020.2991804.
[28] B. Karanayil, M. Ciobotaru, and V. G. Agelidis, “Power flow management of isolated multiport converter for more electric
aircraft,” IEEE Trans. Power Electron., vol. 32, no. 7, pp. 5850–5861, Jul. 2017, doi: 10.1109/TPEL.2016.2614019.
[29] J. Chen, C. Wang, and J. Chen, “Investigation on the selection of electric power system architecture for future more electric
aircraft,” IEEE Trans. Transp. Electrific., vol. 4, no. 2, pp. 563–576, Jun. 2018, doi: 10.1109/TTE.2018.2792332.
[30] “C3D10065A Datasheet.” Cree, Inc., Dec. 2015.
[31] D. Marroqui, A. Garrigos, J. M. Blanes, R. Gutierrez, E. Maset, and F. Iannuzzo, “SiC MOSFET vs SiC/Si Cascode short circuit
robustness benchmark,” Microelectronics Reliability, vol. 100–101, p. 113429, Sep. 2019, doi: 10.1016/j.microrel.2019.113429.
[32] C. Zhao, L. Wang, X. Yang, F. Zhang, and Y. Gan, “Comparative investigation on paralleling suitability for SiC MOSFETs and
SiC/Si Cascode Devices,” IEEE Trans. Ind. Electron., vol. 69, no. 4, pp. 3503–3514, Apr. 2022, doi: 10.1109/TIE.2021.3070519.
[33] S. Martin, T. Kayiranga, Y. Shi, and H. Li, “Comparative study of a 100kW PV WBG inverter using 1200V SiC MOSFET and
JFET cascode devices,” in 2018 IEEE Applied Power Electronics Conference and Exposition (APEC), San Antonio, TX, Mar.
2018, pp. 399–405. doi: 10.1109/APEC.2018.8341042.
[34] B. Vasilev, “Analysis and improvement of the efficiency of frequency converters with pulse width modulation,” IJECE, vol. 9,
no. 4, p. 2314, Aug. 2019, doi: 10.11591/ijece.v9i4.pp2314-2320.
[35] S. Suroso, D. T. Nugroho, A. N. Azis, and T. Noguchi, “Simplified five-level voltage source inverter with level-phase-shifted
carriers based modulation technique,” IJEECS, vol. 13, no. 2, p. 461, Feb. 2019, doi: 10.11591/ijeecs.v13.i2.pp461-468.
[36] E. L. Carvalho, L. H. Meneghetti, E. G. Carati, J. P. da Costa, C. M. de O. Stein, and R. Cardoso, “Asymmetrical pulse-width
modulation strategy for current-fed dual active bridge bidirectional isolated converter applied to energy storage systems,”
Energies, vol. 13, no. 13, p. 3475, Jul. 2020, doi: 10.3390/en13133475.
[37] M. Shahab, S. Wang, and A. K. Junejo, “Improved control strategy for three-phase microgrid management with electric vehicles
using multi objective optimization algorithm,” Energies, vol. 14, no. 4, p. 1146, Feb. 2021, doi: 10.3390/en14041146.
[38] N. A. Rai, M. J. A. Aziz, M. R. Sahid, and S. Md. Ayob, “Bridgeless PFC single ended primary inductance converter in
continuous current mode,” IJPEDS, vol. 10, no. 3, p. 1427, Sep. 2019, doi: 10.11591/ijpeds.v10.i3.pp1427-1436.
[39] J. L. Afonso et al., “A Review on Power Electronics Technologies for Power Quality Improvement,” Energies, vol. 14, no. 24, p.
8585, Dec. 2021, doi: 10.3390/en14248585.

Int J Pow Elec & Dri Syst, Vol. 13, No. 4, December 2022: 1951-1965
Int J Pow Elec & Dri Syst ISSN: 2088-8694  1965

[40] S. Prakash, J. K. Singh, R. K. Behera, and A. Mondal, “Comprehensive analysis of SOGI-PLL based algorithms for single-phase
system,” in 2019 National Power Electronics Conference (NPEC), Tiruchirappalli, India, Dec. 2019, pp. 1–6. doi:
10.1109/NPEC47332.2019.9034724.
[41] N. S. Nise, Control systems engineering, 6th ed. Hoboken, NJ: Wiley, 2011.
[42] E. O. Prado, P. C. Bolsi, H. Confortin Sartori, and J. Renes Pinheiro, “Analytical model for the calculation of losses in power
mosfets for database applications,” REP, vol. 26, no. 04, pp. 388–398, Dec. 2021, doi: 10.18618/REP.2021.4.0023.
[43] C. W. T. McLyman, Transformer and inductor design handbook, 3rd ed., rev.Expanded. New York: Marcel Dekker, 2004.

BIOGRAPHIES OF AUTHORS

Osmar Felipe Alves Eleodoro was born in Alta Floresta, Brazil, in 1999. He is
currently pursuing a M.Sc degree in Electrical Engineering at the Research Center in Power
Electronics (NUPEP) of the Federal University of Uberlândia (UFU). His areas of interest are:
Wide-bandgap semiconductor applications, switched converters, electric vehicles, and digital
control applied to power electronics. He can be contacted at email: [email protected].

Vitor Fonseca Barbosa holds a B.Sc degree from the Federal University of
Triângulo Mineiro (UFTM) and an M.Sc degree from the Federal University of Uberlândia
(UFU) in 2017 and 2020, respectively. He is currently pursuing a Ph.D in Electrical
Engineering at the of the Research Center in Power Electronics (NUPEP) at UFU. His areas
of interest are: active power factor correction, digital signal processing, switched converters,
hybrid rectifiers, DC microgrids and more electric aircraft (MEA). He can be contacted at
email: [email protected].

Gustavo B. Lima was born in Ibiassucê, Brazil, in 1986. Received the B.Sc.,
M.Sc., and Ph.D. degrees in electrical engineering from the Universidade Federal de
Uberlandia, Uberlandia, MG, Brazil, in 2010, 2012, and 2015, respectively. He is currently
with the Research Center in Power Electronics (NUPEP) at UFU and since 2017 he has been a
professor member. His research interests include hybrid rectifiers, digital control applied to
power electronics and power factor correction. He can be contacted at email:
[email protected].

Luiz Carlos Gomes Freitas was born in Uberlândia, Brazil, in 1976. Graduated
in Electrical Engineering with emphasis on Power Systems from the Federal University of
Uberlândia (2001), master’s degree (2003) and Ph.D (2006) in Electrical Engineering with
emphasis on Power Electronics from the same university. In his doctoral thesis he developed
an innovative topological design of a three-phase hybrid rectifier for high power drive
systems. In 2012, he received the 2nd Prize Paper Award of the IEEE-IAS-Industrial
Automation and Control Committee (IACC) for his contribution to the development of hybrid
rectifier structures. In 2008 he joined the faculty of the Federal University of Uberlândia
where he has been working developing teaching and research activities in the area of Power
Electronics and Power Systems. Since 2010 he has been the coordinator of the Research
Center in Power Electronics (NUPEP) at UFU. Since 2013 he has been a researcher
recognized by National Council for Scientific and Technological Development (CNPq) with a
research productivity grant. he has experience in the area of electrical engineering, with an
emphasis on conversion and rectification of electric energy, working on various topics related
to power electronics, electric power quality and renewable energy. He can be contacted at
email: [email protected].

Analysis of SiC and SiC-Cascode MOSFET in the design of power … (Osmar Felipe Alves Eleodoro)