IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO.

2, FEBRUARY 2015

727

Design and Analysis of Class E/F3 Power

Amplifier with Nonlinear Shunt Capacitance
at Nonoptimum Operation
Mohsen Hayati, Akram Sheikhi, and Andrei Grebennikov, Senior Member, IEEE

AbstractThis paper presents analytical expressions for the

Class-E/F3 power amplifier with a nonlinear shunt capacitor for
satisfying the nonoptimum condition at 50% duty ratio. The design parameters for achieving the nonoptimum condition are expressed as a function of the phase shift and dc supply voltage. The
Class-E/F3 amplifier with nonoptimum condition increases one
design degree of freedom compared with the optimum conditions.
Because of the increase in the design degree of freedom, one more
relationship can be specified as a design specification. The following
parameters series reactance, peak switch voltage, output power capability, and maximum operating frequency are presented. Based
on analytical results, an example of the Class E/F3 at operating
frequency of 4 MHz is analyzed. The PSpice simulation and measured results agree with the analytical expressions, which show the
validity of our analytical expressions at zero-voltage switching.
Index TermsMixed mode, nonlinear capacitance, nonoptimum
operation and zero-voltage switching (ZVS), power amplifier.

I. INTRODUCTION
HE nonoptimum operation of the amplifier occurs when
the zero-voltage switching (ZVS) condition is satisfied,
but the zero-voltage-derivative switching (ZDS) is not equal to
zero at the switch turns ON instant. The concept of nonoptimum
operation of the Class-E amplifier was defined in the beginning
of the Class-E history [1][4]. The first research on nonoptimum
operation has been done by Raab [1]. In this paper, the degree
of freedom for the design amplifier was increased, and the ZDS
condition was removed. The mixed-mode power amplifiers are
the good choice for obtaining high-power and high conversion
efficiency. The optimum conditions of the mixed-mode power
amplifier families with a shunt capacitor have been presented
in [5][7]. But the exact analysis on switch mode PAs are not presented. However, all of the Class-E/F amplifier analyses focus
on how to achieve the optimum operation. Many power electronic devices only need to ZVS or zero-current switching (ZCS)
condition [8][20]. In order to reach a Class E power amplifier

Manuscript received December 26, 2013; revised January 19, 2014; accepted
February 13, 2014. Date of publication February 25, 2014; date of current
version October 7, 2014. Recommended for publication by Associate Editor
T. J. Liang.
M. Hayati and A. Sheikhi are with the Department of Electrical Engineering,
Faculty of Engineering, Razi University, Tagh-E-Bostan, Kermanshah-67149,
Iran (e-mail: [email protected]; [email protected]).
A. Grebennikov is with the Bell Laboratories, Alcatel-Lucent, Dublin 15,
Ireland (e-mail: [email protected]).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TPEL.2014.2308280

with good performance, some methods have been suggested and

implemented. The first method employs the push/pull topology
to Class E in order to short odd harmonic like inverse Class F
(demonstrated Class E/F PA). This configuration improved the
amplifier performance such as maximum operating frequency
and caused the reduction of maximum switching voltage to have
maximum output power capability [21], which only in the ZVS
is considered. But the nonoptimum condition in the push/pull
topology is considered with linear shunt capacitance. The second method is harmonic tuning in which a resonance network is
inserted between the drain node and ground. In comparison with
Class E amplifier, the performance of the power amplifier can
be improved at the cost of the complicated circuit design [6].
Besides the aforementioned methods, some Class E amplifiers
consist of resonance networks tuned at nonharmonic frequencies. Nonharmonic resonance network consists of parallel finite
dc feed inductor and parallel drain to source capacitor [22], [23]
and the series parallel resonance network at the drain node [24].
In [25], a parallel LC resonator, tuned to the second harmonic
of the switching frequency, is added into the load network of
the Class-E amplifier, in which the significant improvement in
output power capability is achieved. Up to 10 MHz, the power
amplifier with nonoptimum condition can reach high power
conversion efficiency, so the analysis of the power amplifier at
nonoptimum condition is very important.
In this paper, the analytical expressions of the Class-E/F3
power amplifier with a shunt capacitor at the nonoptimum operation are presented. For the verification of our analytical expressions, design examples of the Class-E/F3 ZVS amplifier are
presented. The design equations, switch, and output waveforms
are obtained as a function of the phase shift and dc supply voltage. By changing the peak switch voltage and the peak switch
current, the phase shift is changed, so these two parameters
can be considered as a design specification. The measurement
and PSpice-simulation results are in good agreement with the
analytical results. Design procedures are given for one example
of Class E/F3 amplifiers for nonoptimum operation. In the example, operating frequency, input dc supply voltage, load resistance, and peak switch voltage are given as design specifications.
In the nominal circuit, only three parameters such as operating
frequency, input dc supply voltage, and load resistance can be
designated. On the contrary, the example procedures presented
here have four given parameters. The additional parameters,
peak switch voltage or peak switch current can be taken into
account in the design procedure. This power amplifier is implemented with the IRF510 MOSFET. From the breakdown voltage

0885-8993 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications standards/publications/rights/index.html for more information.

728

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO. 2, FEBRUARY 2015

The current through the MOSFET device is

iM OS () = ID D In sin(n) IM sin( + )

(5)

and on the other hand, the current of the MOSFET is depicted

as
iM OS () = is () + iC ds () for 0 < < .

(6)

So during the off-state, 0 < < , the current of the

MOSFET is depicted as
Fig. 1.

Cj 0 dvs
iM OS () = iC ds () = 
1 + Vvbsi d

Circuit and model of class-E/Fn power amplifier analyzed in this paper.

point of view, it is proven that the IRF510 MOSFET is a proper

device.
II. CIRCUIT ANALYSIS
The circuit analyzed in this paper is a Class E/F3 amplifier
which shown in Fig. 1. The shunt capacitance of the amplifier
consists only of the MOSFET output parasitic capacitance which
is nonlinear. The derivations of design equations are carried out
under the following assumptions:
1) the inductance of the choke LRFC is large enough to neglect its current ripple;
2) the loaded quality factor Q of the output resonance circuit
is high enough so that the output current can be considered
as a sine wave;
3) the on and off resistance of the MOSFET is zero and
infinite, respectively;
4) the shunt capacitance Cds is entirely composed of the
MOSFET output capacitance. The grading coefficient of
the MOSFET output capacitance is 0.5. The shunt capacitance is described by
Cds = 

Cj 0
1+

vs
Vb i

(1)

where Vbi is the built-in potential, which changes from 0.5

to 0.9 V, vs is the switch voltage, and Cj 0 is the capacitance
at vs (0) = 0;
5) the loaded quality factor Q of the output resonant circuit
is defined as
L
.
(2)
Q=
R
The quality factor Q is high enough to generate a sinusoidal
output current
io = IM sin( + )

(3)

where IM is the amplitude of the output current, = t, is

the angular switching frequency, and is a phase-shift between
the gate-to-source voltage and the output current. The current
through the shunt capacitance produces the switch voltage. The
switch-voltage waveform satisfies the ZVS conditions at the
switch turn-on instant. We assume that the switch for 0 < <
is OFF, so we have
vs () = 0.

(4)

(7)

where, iC ds () is the current through the MOSFET parasitic

capacitance. From (7) and vs (0) = 0, we obtain
 vs

C
 j 0  dvs , for 0 < < .
iM OS ( ) d =
0
0
1 + Vvbsi
(8)
By substituting (6) into (8), we have

[ID D In sin(n) IM sin( + )] d
0

vs

=
0

Cj 0
1+

v s
Vb i

dvs , for 0 < < . (9)

From (9), we obtain

vs () = Vbi [(h() + 1)2 1]

(10)

where
(cos(n) 1)+IM (cos( + ) cos())
.
2Vbi Cj o
(11)
By substituting = and (4) into (10), we obtain

h() =

ID D +

In
n

In
(cos(n) 1) 2IM cos() = 0.
(12)
n
The relative amplitude of the nth harmonic component for
any n can be obtained [7] as
ID D +

4ID D
.
n
By substituting (13) into (12), for n = 3, we have
ID D =

In =

(13)

18
IM cos().
9 2 8

(14)

In power amplifier, the dc-supply power PI

PI = VD D ID D

(15)

and the output power Po is

2
R IM
.
2
The power conversion efficiency is

Po =

2
Po
R IM
=
= 1.
PI
2 ID D VD D

(16)

(17)

HAYATI et al.: DESIGN AND ANALYSIS OF CLASS E/F3 POWER AMPLIFIER

Fig. 2.

729

Normalized input resistance V D D /RID D as a function of .

It is assumed that there is no power losses in the circuit since, in any cases different from optimum Class E with
zero-voltage and zero voltage-derivative conditions, there exists
a nonzero value of current through the drain-source capacitance
at t = , which means that the existing nonzero voltage across
the capacitance should be discharged instantaneously through
the ideal switch with zero resistance.
From (14) and (17), the amplitude of the output current is
36 VD D cos()
9 2 8
R
and the amplitude of the output voltage
IM =

VM = RIM =

36
VD D cos()
9 2 8

and the dc-supply current is


2
18
2VD D cos2 ()
.
ID D =
9 2 8
R

(18)

(19)

(20)

As can be seen from Fig. 2, the normalized input resistance

is a function of the phase shift and it decreased as phase shift
increased. The dc-supply and output power is

2
18
2 VD2 D cos2 ()
PI = Po = ID D VD D =
. (21)
2
9 8
R

Fig. 3. Typical normalized switch voltage (a) for < o p tim u m ,

(b) > o p tim u m , and (c) = o p tim u m .

The shunt capacitance current at the switch turn-on instant is

obtained as
iC ()| = = ID D IM sin(+)In sin(n) = ID D +IM sin()





18
18
VD D
cos()
=2
cos()+sin()
.
9 2 8
R
9 2 8

(22)
The normalized switch voltage for different value of , is
shown in Fig. 3. The switch turns ON with the switch voltage declining and the ZVS condition is satisfied. As can be
seen from Fig. 3(a) and (b), when < optimum and >
optimum , the switch turns OFF and the ZVS condition is satisfied but ZDS condition is not satisfied. But at optimum condition = optimum , the ZVS and ZDS conditions are achieved.
The normalized output current amplitude IM R/VD D , output
power Po R/VD2 D , and shunt capacitance current RiC /VD D

Fig. 4. Normalized output current amplitude IM R/V D D , output power

P o R/V D2 D and shunt capacitance current R iC /V D D as a function of .

at the switch turn-on instant as a function of are shown

in Fig. 4. For nonoptimum operation, the operating range is
/2 < < 0.61 rad. The output power Po is zero at /2.
In this case, the dc input current ID D is zero and the switch voltage waveform is a part of a sinusoidal waveform. The switch

730

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO. 2, FEBRUARY 2015

Fig. 6.
Fig. 5.

voltage slope at the switch turn-on is zero when = 0.61 rad.

In this case, the amplifier achieves the optimum operation and
the product of the output power Po and the load resistance R
reaches the maximum value.
The average value of the switch voltage is equal to the dcsupply voltage VD D
VD D

1
=
2


0

1
vs ()d =
2

and 1 is
Vbi [(h() + 1) 1]d.

9 cos()
Cj 0 R =
(9 2 8)
 2

4 VD D (27 4 198 2 +152) + 2 VD D cos(2 ) 0.5

(27 4 648 2 + 544) + 4 (9 2 8)2 sin2 ()

2 sin(2 )
.
(9 2 8)

(24)
The variation of Cj 0 R as a function of VD D and is shown
in Fig. 5. The maximum value is obtained when VD D is increased. On the other hand, when VD D = 100 V and = 0.83,
the maximum value for Cj 0 R is obtained. As can be seen from
Fig. 1, the voltage across the reactance Lx is expressed as
vL x = Lx

where V1 is the amplitude of vs1 and is expressed as


2
Lx
V1 = VM 1 +
R

(23)
Solving this equation, one obtains the Cj 0 R as a function
of VD D and . Equation (23) has two solutions. One solution
is not suitable for this circuit because Cj 0 R is negative when
/2 < < 0.1. The other solution of (23) for Cj 0 R is
obtained as

Variation of X/R as a function of V D D and .

Variation of Cj 0 R as a function of V D D and .

dio
Lx VM
= Lx IM cos( + ) =
cos( + )
dt
R
for 0 < < 2. (25)

1 = + tan1

vs 1 () = vo () + vL x ()
= VM

Lx VM
cos(+) = V1 sin(+1 ) (26)
sin(+)+
R

Lx
R


.

(28)

The output voltage of the resonant circuit vs1 has only the
fundamental frequency component of the switch voltage vs .
Therefore

1 2
vs () cos( + 1 )d = 0.
(29)
0
From (29), we obtain

vs () cos()d
tan 1 = 0
0 vs () sin()d

Vbi [(h() + 1)2 1] cos()d
= 0
2
0 Vbi [(h() + 1) 1] sin()d

(30)

from (28), the reactance Lx is given by

Lx = X = R tan(1 ).

(31)

In Fig. 6, X/R as a function of and VD D is shown. It can

be seen that the variation of VD D cause a little change in X/R,
but this ratio decreased when increased.
From the definition of the loaded quality factor Q, L is obtained as
L=

The voltage across the resonant circuit for 0 < < 2 is

(27)

RQ
.

(32)

From (31) and (32), the inductor Lr is

Lr = L Lx =

R
[Q tan(1 )].

(33)

HAYATI et al.: DESIGN AND ANALYSIS OF CLASS E/F3 POWER AMPLIFIER

Fig. 7.

Variation of V S M /VD D as a function of V D D and .

731

Fig. 8.

Variation of IS M R as a function of V D D and .

Fig. 9.

Variation of C p as a function of V D D and .

The resonant capacitor is obtained as

C=

1
2 Lr

1
.
R[Q tan(1 )]

(34)

III. OUTPUT POWER CAPABILITY

The definition of output power capability Cp is
Cp =

VD D ID D
=
|Vs m ax |Is m ax ||

1
|V s m a x | |I s m a x |
VD D
ID D

(35)

where Vs m ax and Is m ax are the peak value of voltage and current

at the MOSFET. The peak switch voltage appears in the range
of 0 < < . Therefore, we have
|vs (m ax )|
Vbi
|Vs m ax |
=
=
[(h(m ax ) + 1)2 1]. (36)
VD D
VD D
VD D
In the peak value of the voltage, the derivative value of the
switch voltage is zero. So, from (6) and (8) for n = 3, we
obtained
ID D I3 sin(3 m ax ) IM sin(m ax + ) = 0.

(37)

By substituting (13), (18), and (20) into (37), we have

 
4
9 2 8
sin(m ax + ) = 0. (38)
1
sin(3m ax ) +
3
18
Fig. 7 shows the ratio VS M AX /VD D as a function of and
VD D . This ratio increases from 3.5 to about 3.8 as VD D increases
from 0 to 100 V. On the other hand, the peak switch current
appears during the on-state. So, we have
IS M AX
I3
IM
=1
sin(3)
sin( + ).
ID D
ID D
ID D

(39)

The normalized switch current IS M AX /ID D is function of .

Fig. 8 shows normalized maximum switch current IS M R as
a function of . It can be seen that the normalized maximum
switch current IS M R increase as increases from 1.57 to
0.61 rad. Fig. 9 shows the output power capability Cp as
a function of and VD D . It can be seen that Cp increases
as increases. The other parameter to evaluate the amplifier
performance is maximum operating frequency. This parameter
causes the limitation on all the parameters of the power amplifier.

Fig. 10.

Variation of Rfm a x as a function of V D D and .

As can be seen from Fig. 10, the maximum operating frequency

increases as VD D increases. On the other hand, it is increased
as increase up to 0.61 rad and decreased from 0.61 up to
0.1 rad.

732

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO. 2, FEBRUARY 2015

TABLE I
THE VALUES OF THE ELEMENTS FOR THE NONOPTIMUM DESIGN EXAMPLE

IV. DESIGN EXAMPLE

A. Design Peocessure
The design specifications for the amplifiers are f = 4 MHz,
VD D = 25 V, Pout = 10.7 W, VS M = 98 V, Q = 10, and D =
0.5.
From (36), we have = 0.21 rad. From (24), R is calculated
as 54.58 . From (32), L is obtained 21.72 H. From (31), we
obtained Lx = 1.68 H. From (33), Lr = 20.05 H. From (34),
the resonant capacitor is obtained C = 78.96 pF. The series
resonant circuit C3 L3 has the effect on shaping the transistor
current and voltage waveforms. The initial value for resonant
circuit C3 L3 is obtained as: L3 = 14 H and C3 = 12.56 pF.
The practical transistor is not an ideal switch, so an optimization
process should be done. The parameters before and after optimization is shown in Table I. The inductance of the RF choke
was selected as Lf = 30 H. From (45), IS M AX = 0.62 A could
be obtained. From (23), we obtained ID D = 0.42 A.

Fig. 11. (a) Class E/F3 power amplifier used for the nonoptimum operation.
(b) The photograph of the fabricated power amplifier.
TABLE II
THE VALUES OF THE ELEMENTS AND POWER RELATION FOR THE DESIGN
EXAMPLE AT NONOPTIMUM CONDITION

B. Simulation and Experimental Results

The theoretical results have been verified experimentally using the circuit of Class E/F3 in Fig. 11(a). The photograph of the
fabricated power amplifier is shown in Fig. 11(b). In the PSpice
simulation, the device model of IRF510 from the vendors website [26] is used. It is driven by the gate-driver IC 1XDN414P,
its dc supply passes through two parallel capacitors 100 F
and 10 nF. The transistor is mounted directly to a backside aluminum heatsink. The equivalent series resistance (ESR) values
of passive elements were measured by a HP-4284A, and the on
resistance of the IRF510 was obtained from PSpice model.
The following results were obtained: LRFC = 30 H (ferrite
core inductor with air gap), RLRFC = 0.2 (dc parasitic resistance), C3 = 11.89 pF (silver mica capacitors), L3 = 14.1 H
(ferrite core inductor with air gap), RL3 = 0.116 (parasitic
resistance at 12 MHz) C = 78.96 pF (silver mica capacitors),
L = 21.72 H (ferrite core inductor with air gap), RL =
44.67 m (parasitic resistance measured at 4 MHz), and R
= 54.7 (resistance of load network). The power losses of all
capacitors were negligible. A Tektronix TDS3014B digital oscilloscope was used for recording voltage waveform. The output
power POut can be obtained based on the dc power PD C , the
transistor losses PT , the supply current ID C , the currents IC3
and IOut , and the parasitic resistances RLRFC , RL, and RL3 .

The output power is obtained as

POut = PDC PL 3 PL PT
= 10.7 W 0.02 W 0.5 W 0.282 W = 10 W.
The power gain is obtained as

G = 10 log10

Po
Pin


(40)

HAYATI et al.: DESIGN AND ANALYSIS OF CLASS E/F3 POWER AMPLIFIER

733

periments, the Class-E/F3 ZVS amplifier achieves 95.93% and

93.45% power conversion efficiency at 4 MHz and output power
10.3 and 10 W, respectively. We will address the analysis of the
Class-E/Fn amplifier for any n with nonlinear shunt capacitance
in the future works.
REFERENCES

Fig. 12. Waveforms obtained from (a) the PSpice Simulation, and (b) the
experimental results for the nonoptimum operation.

where Pin and Pout are the input and output power, respectively.
Also, the total harmonic distortion (THD) is

THD =

n =2

2
Von

Vo1

(41)

where Von is a root-mean-square value of the nth harmonic

in the output voltage vo . The measured values of the power
gain and THD are 25.38 dB and 4.35%, respectively. The value
of the elements, and all the powers are shown in Table II. The
waveform obtained from Pspice and measured results are shown
in Fig. 12(a) and (b), respectively.
V. CONCLUSION
This paper has introduced analytical expressions of the
Class-E/F3 power amplifier with shunt capacitance for the
nonoptimum operation. The design values for achieving the
nonoptimum condition are expressed as a function of the phase
shift between the input and output voltage and dc-supply voltage. The Class-E/F3 amplifier with nonoptimum condition increases one design degree of freedom. Because of the increase
in the design degree of freedom, peak switch voltage and current can be specified as a design specification. This paper has
given the circuit design example based on our proposed design
expression by specifying the peak switch voltage. The measurement and PSpice simulation results are in agreement with the
analytical expressions quantitatively, which show the validity
of our analytical expressions. In the circuit simulation and ex-

[1] F. H. Raab, Suboptimum operation of Class-E RF power amplifiers, in

Proc. RF Technol. Expo., Santa Clara, CA, USA, Feb. 1989, pp. 8598.
[2] T. Suetsugu and M. K. Kazimierczuk, Analysis of subnominal operation
of Class-E amplifier, presented at the IEEE Midwest Symp. Circuits
Syst., Cairo, Egypt, Dec. 2003.
[3] T. Suetsugu and M. K. Kazimierczuk, Design procedure of class-E amplifier for off-nominal operation at 50% duty ratio, IEEE Trans. Circuits
Syst. I, Reg. Papers, vol. 53, no. 7, pp. 14681476, Jul. 2006.
[4] T. Suetsugu and M. K. Kazimierczuk, Design equations for suboptimum
operation of class E amplifier with nonlinear shunt capacitance, in Proc.
IEEE Midwest Symp. Circuits Syst., vol. 5, Vancouver, Canada, May 2004,
pp. 560563.
[5] Z. Kaczmarczyk, High-efficiency class E, E/F, and E/F inverters, IEEE
Trans. Ind. Electron., vol. 53, no. 10, pp. 15841593, Oct. 2006.
[6] F. You, S. He, X. Tang, and X. Deng, High-efficiency single-ended
Class-E/F2 power amplifier with finite dc feed inductor, IEEE Trans.
Microw. Theory Tech., vol. MTT-58, no. 1, pp. 3240, Jan. 2010.
[7] A. Grebennikov, High-efficiency class E/F lumped and transmissionline power amplifiers, IEEE Trans. Microw. Theory Tech., vol. 59, no. 6,
pp. 15791588, Jun. 2011.
[8] B. Yuan, X. Yang, D. Li, Y. Pei, J. Duan, and J. Zhai, A current-fed multi
resonant converter with low circulating energy and zero-current switching
for high step-up power conversion, IEEE Trans. Power Electron., vol. 26,
no. 6, pp. 16131619, Jun. 2011.
[9] Y.-T. Chen, S. Shiu, and R. Liang, Analysis and design of a zero-voltage
switching and zero-current-switching interleaved boost converter, IEEE
Trans. Power Electron., vol. 27, no. 1, pp. 161173, Jan. 2012.
[10] H. Wang, H. Chung, and A. Ioinovici, A new concept of high-voltage
DCDC conversion using asymmetric voltage distribution on the switch
pairs and hybrid ZVSZCS scheme, IEEE Trans. Power Electron.,
vol. 27, no. 5, pp. 22422259, May 2012.
[11] T.-H. Hsia, H.-Y. Tsai, D. Chen, M. Lee, and C.-S. Huang, Interleaved
active-clamping converter with ZVS/ZCS features, IEEE Trans. Power
Electron., vol. 26, no. 1, pp. 2937, Jan. 2011.
[12] A. Mousavi, P. Das, and G. Moschopoulos, A comparative study of a
new ZCS DCDC full-bridge boost converter with a ZVS active-clamp
converter, IEEE Trans. Power Electron., vol. 27, no. 3, pp. 13471358,
Mar. 2012.
[13] B-R. Lin and P.-J. Cheng, New ZVS DCDC Converter With SeriesConnected Transformers to Balance the Output Currents, IEEE Trans.
Power Electron., vol. 29, no. 1, pp. 246255, Jan. 2014.
[14] A. Safaee, D. Yazdani, A. Bakhshai, and P. K. Jain, Multiblock soft
switched bidirectional ACAC converter using a single loss-less active
snubber block, IEEE Trans. Power Electron., vol. 27, no. 5, pp. 2260
2272, May 2012.
[15] M. Acar, A. J. Annema, and B. Nauta, Generalized design equations for
Class-E power amplifiers with finite DC feed inductance, in Proc. 36th
Eur. Microw. Conf., Manchester, U.K., Sep. 2006, pp. 13021305.
[16] D. K. Choi and S. I. Long, Finite DC feed inductor in Class E power
amplifiersA simplified approach, in Proc. IEEE MTT-S Int. Microw.
Symp. Dig., 2002, pp. 16431646.
[17] S. Bin, Z. Junming, and L. Zhengyu, Totem-pole boost bridgeless PFC
rectifier with simple zero-current detection and full-range ZVS Operating
at the Boundary of DCM/CCM, IEEE Trans. Power Electron., vol. 26,
no. 2, pp. 427435, Feb. 2011.
[18] W. Li, P. Li, H. Yang, and X. He, Three-level forwardflyback phase-shift
ZVS converter with integrated series-connected coupled inductors, IEEE
Trans. Power Electron, vol. 27, no. 6, pp. 28462856, Jun. 2012.
[19] R. Beiranvand, B. Rashidian, M. R. Zolghadri, and S. M. H. Alavi, Optimizing the normalized dead-time and maximum switching frequency
of a wide-adjustable-range llc resonant converter, IEEE Trans. Power
Electron., vol. 26, no. 2, pp. 462472, Feb. 2011.
[20] R. Beiranvand, B. Rashidian, M. R. Zolghadri, and S. M. H. Alavi, A
design procedure for optimizing the lc resonant converter as a wide output range voltage source, IEEE Trans. Power Electron., vol. 27, no. 8,
pp. 37493763, Aug. 2012.

734

[21] S. D. Kee, I. Aoki, A. Hajimiri, and D. Rutledge, The class-E/F family

of ZVS switching amplifiers, IEEE Trans. Microw. Theory Tech., vol. 51,
no. 6, pp. 16771690, Jun. 2003.
[22] G. H. Smith and R. E. Zulinski, An exact analysis of Class-E amplifiers
with finite DC-feed inductance at any output-Q, IEEE Trans. Circuits
Syst., vol. 37, no. 4, pp. 530534, Apr. 1990.
[23] M. Acar, A. J. Annema, and B. Nauta, Analytical design equations for
Class-E power amplifiers, IEEE Trans. Circuits Syst. I, Reg. Papers,
vol. 54, no. 12, pp. 27062717, Dec. 2007.
[24] F. You, S. He, X. Tang, and T. Cao, Performance study of a
Class-E power amplifier with tuned series-parallel resonance network,
IEEE Trans. Microw. Theory Tech., vol. 56, no. 10, pp. 21902200, Oct.
2008.
[25] A. Mediano and N. O. Sokal, A Class-E RF Power amplifier with a flattop transistor-voltage waveform, IEEE Trans. Power Electron., vol. 28,
no. 11, pp. 52155221, Nov. 2013.
[26] International Rectifier. (2011, Jun.). [Online]. Available: http://www.irf.
com/productinfo/models

Mohsen Hayati received the B.E. degree in electronics and communication engineering from Nagarjuna University, Guntur, India, in 1985, and the M.E.
and Ph.D. degrees in electronics engineering from
Delhi University, Delhi, India, in 1987 and 1992,
respectively.
He joined the Department of Electrical Engineering, Razi University, Kermanshah, Iran, as an Assistant Professor in 1993, and currently, as an Associate
Professor. He has published more than 140 papers in
international and domestic journals and conferences.
His current research interests include microwave and millimeter wave devices
and circuits, application of computational intelligence, artificial neural networks,
fuzzy systems, neuro-fuzzy systems, electronic circuit synthesis, and modeling
and simulations.

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 30, NO. 2, FEBRUARY 2015

Akram Sheikhi received the B.E. degree from

Shariaty University, Tehran, Iran, in 2007, and the
M.Sc. degree in electronics engineering from the Department of Electrical Engineering, Razi University,
Kermanshah, Iran, in 2010. She is currently working
toward the Ph.D. degree in electronics engineering in
the Razi University,.
Her research interests include microwave filter, high-frequency high-efficiency switching mixedmode power amplifiers, and numerical analysis of
nonlinear circuits.

Andrei Grebennikov (M99SM04) received the

Dipl. Eng. degree in radio electronics from Moscow
Institute of Physics and Technology, Moscow,
Russia, in 1980, and the Ph.D. degree in radio engineering from Moscow Technical University of Communications and Informatics, Moscow, Russia, in
1991.
He obtained a long-term academic and industrial experience working with the Moscow
Technical University of Communications and Informatics; the Institute of Microelectronics, Singapore; M/A-COM, Cork, Ireland; Infineon Technologies, Munich,
Germany, and Linz, Austria; and Bell Labs, Alcatel-Lucent, Dublin, Ireland,
as an Engineer, Researcher, Lecturer, and Educator. He has lectured as a Guest
Professor with the University of Linz, Linz, Austria, and presented short courses
and tutorials as an Invited Speaker at the IEEE Microwave Theory and Techniques Society (MTT-S) International Microwave Symposia (IMS), European
and Asia-Pacific Microwave Conferences, the Institute of Microelectronics,
Singapore, Motorola Design Centre, Penang, Malaysia, the Tomsk State University of Control Systems and Radio electronics, Tomsk, Russia, and the Aachen
Technical University, Aachen, Germany. He is an author and coauthor of more
than 100 papers, holds 30 European and US patents and patent applications, and
authored six books dedicated to RF and microwave circuit design.