280
2, FEBRUARY 2010
A Wide-Input–Wide-Output (WIWO)
DC–DC Converter
Hao Cheng, Keyue Ma Smedley, Fellow, IEEE, and Alexander Abramovitz, Member, IEEE
Abstract—This paper presents a new wide-input–wide-output
dc–dc converter, which is an integration of buck and boost convert-
ers via a tapped inductor. Coherent transition between step-down
and step-up modes is achieved by a proper control scheme. This
paper presents theoretical concepts and experimental results.
Index Terms—Boost, buck, coupled inductors, energy recovering
snubber, wide step-down, wide step-up, wide-input–wide-output
(WIWO) dc–dc converter.
I. INTRODUCTION
HE BUCK, boost, buck–boost, and Cúk converters are the
T four basic dc–dc nonisolating converters that have found
wide applications in industry. The buck converter can step down
the dc voltage, whereas the boost converter is capable to per-
form a step-up function. In applications where both step-up
and step-down conversion ratios are required, the buck–boost
and Cúk converters can be used. Simplicity and robustness are
among the advantages of the buck–boost converter. However,
the pulsating input and output currents cause high conduction
losses, and thus, impair the efficiency of buck–boost. Further-
more, the buck–boost converter uses the inductor to store the
energy from the input source, and then, release the stored energy
to the output. For this reason, the magnetic components of buck–
boost are subjected to a significant stress. These disadvantages
limit the applications of the buck–boost converter mainly to low
power level. The isolated version of buck–boost, referred to as
the flyback converter, can achieve greater step-up or step-down
conversion ratio utilizing a transformer, possibly, with multiple
outputs. As compared with the buck–boost converter, the Cúk
converter has higher efficiency and smaller ripples in input and
output currents. A significant improvement of the Cúk converter
performance can be achieved by applying the zero ripple con-
cept. The Cúk converter can be found in many high-performance
power applications.
In theory buck and boost converters can generate almost any
voltage, in practice, the output voltage range is limited by com-
ponent stresses that increase at the extreme duty cycle. Conse-
Manuscript received November 7, 2008; revised January 20, 2009. Current
version published February 12, 2010. Recommended for publication by
Associate Editor J. Antenor.
H. Cheng is with the Department of Electrical Engineering and Computer
Science (EECS), University of California, Irvine (UCI), Irvine, CA 92617 USA
(e-mail: hcheng1@uci.edu).
K. M. Smedley is with the Department of Electrical and Computer Engi-
neering (ECE), University of California, Irvine (UCI), Irvine, CA 92697 USA
(e-mail: ksmedley@uci.edu).
A. Abramovitz is with the Department of Electrical Engineering (EE), Sami
Shamoon College of Engineering (SCE), Beer-Sheva 84100, Israel (e-mail:
aabramov@eng.uci.edu).
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.2009.2025375
0885-8993/$26.00 © 2010 IEEE
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO.
quently, buck converter losses mount at low duty cycle, whereas
boost converter efficiency deteriorates when the duty cycle tends
to unity. Accordingly, voltage conversion range of the buck con-
verter below 0.1–0.15 becomes impractical whereas that of the
boost converters’ is limited to below 8–10. Additional prob-
lems associated with narrow duty cycle are caused by MOS-
FET drivers rise and fall times as well as pulsewidth-modulated
(PWM) controllers that have maximum pulsewidth limitations.
These problems become even more severe at higher voltages
and higher frequencies.
Introducing a transformer helps attaining large step-up or
step-down voltage conversion ratio. Transformers’ turn ratio
should be chosen as to provide the desired voltage gain while
keeping the duty cycle within a reasonable range for higher
efficiency. The transformer, however, brings in a whole new
set of problems associated with the magnetizing and leakage
inductances, which cause voltage spikes and ringing, increased
core and cooper losses as well as increased volume and cost.
In a quest for converters with wide conversion range, quite a
few authors proposed using converters with nonlinear charac-
teristics. Single-transistor converter topologies, with quadratic
conversion ratios, were proposed in [1] and demonstrated
large step-down conversion ratio. This method has successfully
achieved wide conversion range in the step done direction. A
different approach to obtain wide conversion range utilizing
coupled inductors was proposed in [2]. With only minor modi-
fication of the tapped-inductor buck, [2] shows low component
count and solves the gate-drive problem by exchanging the po-
sition of the second winding and the top switch. The problem
of a high turn-OFF voltage spike on the top switch was solved
by applying a lossless clamp circuit. Due to the coupled induc-
tor action, the converter demonstrated high step-down dc–dc
conversion ratio, whereas the converter’s efficiency was im-
proved by the extended duty cycle. A tapped-inductor buck
with soft switching was introduced in [3]. Derivations of the
tapped-inductor buck were also suggested in [4] and [5]. An-
other modification of the tapped-buck converter was realized
in [6] for power factor correction (PFC) application. With the
addition of a line-frequency-commutated switch and a diode,
both flyback and buck characteristics were achieved and large
step-down was demonstrated.
Some applications, especially battery-operated equipment,
require high voltage boosting. To attain very large voltage step-
up, cascaded boost converters that implement the output volt-
age increasing in geometric progression were introduced in [7].
These converters effectively enhance the voltage transfer ra-
tio; however, their circuits are quite complex. In comparison,
tapped-inductor boost converters proposed in [8] and [9] at-
tain a comparable voltage step-up preserving relative circuit
CHENG et al.: WIDE-INPUT–WIDE-OUTPUT (WIWO) DC–DC CONVERTER
Fig. 1. Buck-derived converters with tapped inductors.
simplicity. In [10], the boost converter output terminal and fly-
back converter output terminal are connected in series to in-
crease the output voltage gain with the coupled inductor. The
boost converter also functions as an active clamp circuit to re-
cycle the snubber energy.
This paper proposes a new wide-input–wide-output (WIWO)
dc–dc converter. The new converter is an integration of buck
and boost converters via a tapped inductor. By applying proper
control to the two active switches, the converter exhibits both
buck and boost features [11], [12]. Section II discusses the basic
switching converters with tapped inductors, and offers motiva-
tion and guidelines to the synthesis of the new switching con-
verter. Section III presents the topology of the proposed WIWO
dc–dc converter and the required control scheme. The operating
principle is described in detail providing the steady-state (dc)
and dynamic (ac) models as well. The theoretical derivations
were verified experimentally and reported in Section IV. Modi-
fications and additional applications are discussed in Section V.
Conclusions are given in Section VI.
II. MOTIVATION IN THE SEARCH FOR NEW SWITCHING
CONVERTER TOPOLOGY
The basic buck and boost converters can be transformed into
a number of new topologies by bringing in the tapped induc-
tor. The proposed tapped-inductor buck-derived converters are
shown in Fig. 1, with their corresponding voltage conversion
ratios plotted in Fig. 2. The proposed tapped-inductor boost-
derived topologies and their corresponding voltage conversion
ratios are given in Figs. 3 and 4. Here, D is the duty ratio of
switch S, M is the voltage conversion ratio, and n is the turn
ratio of the tapped inductors, which is defined as n = n2 : n1 .
As the turn ratio n tends to infinity, the conversion ratio of the
buck-derived converters approach the characteristic of a basic
buck topology. On the other hand, as the turn ratio n goes to
zero, the conversion ratio of the boost-derived converters ap-
proach the characteristic of a basic boost topology. Inspection
of the conversion ratio plots, as given in Fig. 1(a), reveals that the
proposed buck-derived converter achieves wider voltage step-
down than a basic buck converter. Also, by examining Fig. 3(a),
it becomes evident that the suggested boost-derived converter
attains a wider voltage step-up than a basic boost converter.
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
281
The converter topologies shown in Figs. 1(a) and 3(a) are
strikingly similar. The idea proposed here is that these two
topologies may be combined to form a new two-switch topol-
ogy, with an extended conversion range. Same conclusion can
be reached comparing the converters given in Figs. 1(c) and 3(c).
The proposed WIWO range converter topology is described in
the next section.
III. WIWO DC–DC CONVERTER
A. Proposed WIWO DC–DC Converter Topology
The proposed WIWO dc–dc converter is illustrated in Fig. 5.
The converter is comprised of two active switches S1 and S2,
tapped inductors L1 and L2 with turns ratio n = n2 : n1 , diode
D, and capacitive output filter C.
Specifically, note that the tapped inductor in Figs. 1 and 3 is
reconfigured into a pair of coupled inductors in Fig. 5. Being
equivalent electrically, this reconfiguration is beneficial from a
practical point of view. In Fig. 5, S1 and S2 are connected to
a common junction or midpoint. The midpoint is periodically
switched by S1 to ground, which allows recharging the bootstrap
power supply and reliable operation of the flying driver of the
top switch S2. Consequently, a standard half-bridge driver chip
can be used with the low-side driver operating the bottom switch
S1 and the bootstrap high-side driver activating the top switch
S2.
WIWO can operate either in the step-down or the buck mode
or in the step-up or the boost mode. To operate the WIWO
in the buck mode, the switch S1 is assigned a high-frequency
switching signal with a predetermined duty cycle D, whereas
S2 is switched complementarily to S1. The diode D is kept ON
by the inductor L2 current, which is assumed to be continuous.
To operate WIWO in the boost mode, the controller keeps
S2 switch continuously ON and issues the required duty cycle
signal for the S1 switch. Thus, the diode D is forced to switch
on and off complementarily to S1.
In both modes, the capacitor C filters the pulsating current
and provides a smoothed output voltage for the load R.
B. Control Scheme
For the proper operation of WIWO, a modified PWM control
circuitry is required. The implementation is not unique. One
possible realization of the modulator is shown in Fig. 6. Here, a
window comparator is employed to derive the required switch-
ing signals for S1 and S2 by comparing the sawtooth ramp with
amplitude of Vm to the two control voltages VC and VC
. The
control voltage VC for the upper comparator is delivered by
an external source, whereas the lower comparator input signal
VC
is derived by the PWM circuitry, downshifting the control
voltage VC by Vm : VC
= VC − Vm . The relationship between
the control voltage VC and the sawtooth ramp amplitude Vm
can be expressed by means of a variable m as VC = mVm .
WIWO operates in the buck mode when 0 < VC < Vm , i.e.,
when 0 ≤ m < 1. Here, the upper comparator generates the
required duty cycle for the S2 switch, whereas the lower com-
parator is in “1” state and commands the NAND gate to provide
282
Fig. 2. Voltage conversion ratio of buck-derived converters with tapped inductors. (a) 0 < n < ∞. (b) 0 < n < ∞. (c) n > 1. (d) n > 1.
Fig. 3. Boost-derived converters with tapped inductors.
the complimentary duty cycle for the S1 switch. Therefore,
WIWO operates similarly to a synchronous buck converter. On
the other hand, for Vm < VC < 2Vm , or 1 ≤ m < 2, the up-
per comparator is in “1” state and keeps S2 continuously ON,
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 2, FEBRUARY 2010
whereas the lower comparator and the NAND gate provide the
required duty cycle for the S1 switch. Thus, the converter enters
the boost mode.
C. Operating Principle of the WIWO Converter
In the following, the steady-state operation of the proposed
WIWO converter is described. The analysis is performed as-
suming that the circuit is comprised of ideal components. The
coupling coefficient of the tapped inductor is assumed to be
unity. Under continuous inductor current (CCM) condition, the
proposed WIWO converter exhibits four topological states, as
shown in Fig. 7. Here, the large output filter capacitor is re-
placed by an ideal voltage source. The waveforms and timing of
WIWO for both buck and boost modes are illustrated in Fig. 8.
1) Buck Mode: State 1 (t0 ≤ t < t1 ) is the buck-mode
charging state [see Figs. 7(a) and 8(a)]. Here, the switch S2
is turned on and S1 is turned off. The diode D conducts and the
coupled inductors L1 and L2 are charged. The energy is also
transferred from dc source to load.
CHENG et al.: WIDE-INPUT–WIDE-OUTPUT (WIWO) DC–DC CONVERTER 283

Fig. 4. Voltage conversion ratio of boost-derived converters with tapped inductors. (a) 0 < n < ∞. (b) 0 < n < ∞. (c) 0 < n < 1. (d) 0 < n < 1.

Fig. 5. WIWO dc–dc converter topology.

State 2 (t1 ≤ t ≤ t2 ) is the buck-mode discharging state [see

Figs. 7(b) and 8(a)]. Here, the switch S2 is turned off also cutting
off the current in the L1 winding, whereas S1 is turned on and
the diode D conducts L2 current to the load.
2) Boost Mode: State 3 (t
0 ≤ t < t
1 ) is the boost-mode
charging state [see Figs. 7(c) and 8(b)]. Here, the switches S1
Fig. 6. Proposed WIWO dc–dc converter and PWM control circuitry.
and S2 are turned on charging the L1 inductor. The diode D
is cut off by the negative voltage induced in L2 winding. The
output voltage is supported by the capacitor C.

Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
284
Fig. 7. Four topological states of the WIWO converter. (a) Buck-mode charging state. (b) Buck-mode discharging state. (c) Boost-mode charging state.
(d) Boost-mode discharging state.
Fig. 8. Waveforms of the WIWO dc–dc converter. (a) Buck mode. (b) Boost
mode.
State 4 (t
1 ≤ t ≤ t
2 ) is the boost-mode discharging state [see
Figs. 7(d) and 8(b)]. Here, the switch S2 is still ON whereas S1 is
turned off. Both windings L1 and L2 conduct through the diode
D and discharge the stored energy to the output.
D. Steady-State Analysis
The steady-state models of the proposed WIWO converter
are shown in Fig. 7. These models preserve the tapped-inductor
symbol. More suitable for analysis purposes, however, are the
models of Fig. 9. Here, the role of the magnetizing inductance
Lm is clearly shown. The detailed analysis was carried out
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 2, FEBRUARY 2010
in [11] using state-space averaging technique. WIWO voltage
conversion ratio, output voltage ripple, voltage stresses, etc.,
were obtained. The characteristics of WIWO are summarized
in Table I for a general case of n and separately for the special
case of n = 1.
WIWO voltage transfer characteristics M (n, m) are plotted in
Fig. 10. Clearly, for n = 1, the voltage transfer ratio is smooth
at the vicinity of the buck to boost switchover point m = 1,
whereas for other values of n, the curves exhibit a slope change.
This statement can be verified analytically by calculating the
derivatives of M (m) at m = 1. Using the expressions for voltage
conversion ratio given in Table I, the result is ((n + 1)/n) for the
buck mode and (n + 1) for the boost mode. Obviously, the slope
of WIWO dc–dc characteristic becomes continuous for n = 1.
Table I also presents the line-to-output and control-to-output
transfer functions. The small-signal transfer functions of the
WIWO converter were derived by linearizing the state-space
equations around the operating point [11]. The line-to-output
and control-to-output transfer functions reveal strong depen-
dence on the operating point and a right-half-plane (RHP)
zero. This is also the case in other tapped-inductor topolo-
gies [13], [14]. These characteristics make the WIWO com-
pensation network design somewhat difficult.
IV. EXPERIMENTAL RESULTS
A 100-W prototype WIWO converter was designed for input
voltage range of 12–48 Vdc and a constant output voltage of
28 Vdc . The turn ratio of the tapped inductor was set to n =
1 with a total inductance of 400 µH. The switching frequency
of 200 kHz was chosen. The tapped inductors were wound on
C058548A2 toroidal powder core, chosen for its low leakage,
with 50 turns of AWG20 wire for both windings. The design
yielded 400 µH inductance with only 560 nH leakage induc-
tance. Two FDD2572 MOSFETs were paralleled to comprise
the top switch and two IRFR3518 were used for the low switch
providing low Rds−ON and low gate capacitance. Schottky diode
CHENG et al.: WIDE-INPUT–WIDE-OUTPUT (WIWO) DC–DC CONVERTER 285

TABLE I
CHARACTERISTICS OF WIWO DC–DC CONVERTER AND CHARACTERISTICS FOR n = 1

Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
286
Fig. 9. Switched circuit models. (a) State 1. (b) State 2. (c) State 3. (d) State 4.
Fig. 10. Voltage transfer characteristics M (n, m) of the WIWO dc–dc
converter.
20CTQ150 was selected due to superior reverse recovery char-
acteristics.
Experimental waveforms of WIWO converting 48 V input
to 28 V output (buck mode) are shown in Fig. 11. In the buck
mode, S2 is the leading switch, gated by the duty cycle command
shown as the bottom trace in Fig. 11, whereas the bottom switch
S1 is switched complementarily, similarly to a synchronous
converter. Switch voltages (see Fig. 6 for definition) are shown
as top two waveforms in Fig. 11. The middle traces show the
winding currents. These were measured by ac probe, so only
the ripple components could be observed. As could be seen, as
the S2 switch conducts, both windings carry the same current.
At the S2 is turned off, the input current ceases whereas the
output current is doubled in amplitude, consistent with WIWO
models in Fig. 9(a) and (b). The ramp portion of the current
is hardly noticeable due to the relatively high frequency and
sufficiently large inductance value. The leakage inductance of
L1 developed a turn-OFF voltage spike across S1 that is smoothed
by the snubber circuitry. The snubber is used to clamp the voltage
spike, as described later.
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 2, FEBRUARY 2010
Fig. 11. Experimental waveforms of the WIWO converter in the buck mode
(see Fig. 6 for designation of variables). Top trace: drain voltage V 1 of S1
switch (50 V/division, 2 µs/division); second top trace: drain voltage V 2
of S2 switch (50 V/division, 2 µs/division); middle trace: input current Ii
(0.2 A/division, 2 µs/division); second bottom trace: output current Io (0.2
A/division, 2 µs/division); bottom trace: S2 switch gating voltage (20 V/ divi-
sion, 2 µs/division).
The experimental waveforms of WIWO in the boost mode
with 12 V input and 28 V output, under full-load condition,
are shown in Fig. 12. To supply the power requirements of the
load at lower input voltage range, WIWO calls for greater input
current, and therefore, turn-OFF voltage spike on S1 is observed.
In the boost mode, the S1 switch is the leading switch that
is issued the duty cycle command, shown as the bottom trace
in Fig. 12. Since in the buck mode the S2 switch is constantly
ON, the drain voltage of S2 and the drain voltage of S1 are
almost identical. The winding currents were measured by a
high-frequency ac probe, and therefore, only ac current compo-
nents are shown as two middle traces in Fig. 12. As S1 switch
conducts, the input winding carries the input current and is
charging, whereas the output current is cut off. As the S1 switch
CHENG et al.: WIDE-INPUT–WIDE-OUTPUT (WIWO) DC–DC CONVERTER 287

Fig. 12. Experimental waveforms of WIWO in the boost mode (see Fig. 6 for
designation of variables). Top trace: drain voltage of S1 switch (20 V/division,
2 µs/division); second top trace: drain voltage of S2 switch (20 V/division,
2 µs/division); middle trace: input current Ii (0.5 A/division, 2 µs/division);
second bottom trace: output current Io (0.5 A/division, 2 µs/division); bottom
trace: S1 switch gating voltage (20 V/division, 2 µs/division).

Fig. 13. Comparison of K with K c rit for n = 1.

is cut off, both windings carry the same current and are dis-
charging into the output capacitor and feeding the load. For this
reason, the currents ripple components appear in antiphase, as
predicted by WIWO models in Fig. 9(c) and (d). Also could
be seen is the snubber circuit resonant discharge as the snubber
recycles the stored energy.
With decreased load, the converter enters the discontinuous
conduction mode (DCM). To measure the tendency of the con-
verter to operate in DCM, the parameter K = (2Lm /RTs ) is
defined as suggested in [15]. The critical value of K for n = 1
is compared with K = 2, 0.2, 0.02 in Fig. 13. The experimen-
tal voltage conversion ratio M as function of m for different
values of K plotted on top of the theoretical curve is given in
Fig. 14(a)–(c). Due to the parasitic resistances in the circuit, the
experimental voltage conversion ratio M is slightly lower than
theoretical prediction. For very same reason, the experimental
M cannot become infinite and drops as m approaches 2. A
narrow buck- to boost-mode transition can be observed on the
WIWO characteristic in the vicinity of m = 1. The conversion
Fig. 14. Comparison of the experimental and theoretical voltage conversion
ratio under different loading conditions. (a) K = 2. (b) K = 0.2. (c) K = 0.02.
ratio in DCM is higher than that in CCM, as shown in Fig. 14(b)
and (c).
The efficiency of the experimental WIWO dc–dc converter
for different dc input voltages versus the load current is plotted
in Fig. 15. The output voltage was kept at the nominal value
of 28 Vdc . No attempt was made to optimize the preliminary
design, still the converter demonstrated high efficiency.
V. APPLICATIONS
A continuously conducting diode D has a considerable
forward voltage drop. This is not desirable for low-output-
voltage applications. The voltage drop can be reduced using

Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
288
Fig. 15. Experimental WIWO converter efficiency.
Fig. 16. WIWO dc–dc converter with the synchronous rectifier.
Fig. 17. Bidirectional WIWO dc–dc converter.
a synchronous rectifier with low Rds -ON instead of the diode,
as shown in Fig. 16.
Interchanging the position of the inductor L2 and switch S3,
as shown in Fig. 17, the WIWO topology becomes symmet-
rical. This also allows driving the top switch S3 with another
flying driver. An additional advantage of the circuit in Fig. 17
is the ability to sustain a bidirectional power flow. The direc-
tion of the power flow can be controlled applying a single-pole
double-throw switch, which may be controlled manually or au-
tomatically, as illustrated in Fig. 17. This WIWO can be used in
a battery charging and discharging application. With the switch
in position 1, the power flows from the left port to the right
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 25, NO. 2, FEBRUARY 2010
Fig. 18. WIWO PFC ac–dc converter.
Fig. 19. Energy recovering snubber for WIWO power stage.
port, whereas with the switch in position 2, the power flows in
a reverse direction from the right port to the left port.
The WIWO dc–dc converter can also be used for PFC appli-
cation (see Fig. 18). Here, a sinusoidal line voltage is fed into
the rectifier input. The WIWO dc–dc converter can accept the
rectified voltage and directly produce the required low dc out-
put. With the line voltage greater than the output, the converter
works in the buck mode. As the line drops below the output
voltage, WIWO enters the boost mode.
VI. ENERGY RECOVERY SNUBBER
Since the WIWO operates a coupled inductor, the energy
stored in the leakage inductances becomes a problem to deal
with. Besides increased switching losses, discharge of the leak-
age inductance energy causes oscillations and increased voltage
spikes across the switches. The resulting voltage stress becomes
intolerable at higher voltages and higher power. If not snubbed,
overvoltage breakdown of the MOSFET devices may occur.
The proposed lossless snubber is comprised of a snubber
capacitor CS and a pair of fast diodes DS1 and DS2 . The snubber
is fitted to WIWO, as shown in Fig. 19. The snubber is effective
both in buck mode and in boost mode.
Detailed description of the snubber operation is out of scope
of this paper; in brief, however, the operation is as follows. With
WIWO in the buck mode, at the instant when the S2 switch is
turned off, the snubber diode DS1 conducts L1 leakage current
and allows the stored energy to be discharged into the snubber
capacitance CS and to the output of the circuit. This takes one-
half resonant cycle dictated by the leakage inductance and the
snubber capacitance. CS will remain charged until the S2 switch
is turned ON again at the onset of the subsequent switching cycle.
With S2 turned ON, the energy stored in the snubber capacitor
is discharged into L2 winding via DS2 and recycled.
CHENG et al.: WIDE-INPUT–WIDE-OUTPUT (WIWO) DC–DC CONVERTER
In the boost mode, the snubber operation is similar. However,
here, S1 interrupts the current and is subject to the voltage spike
while S2 switch is constantly ON with zero voltage VD S 2 across.
VII. CONCLUSION
This paper has presented a new WIWO dc–dc converter,
which is an integration of buck and boost converters with cou-
pled inductors. The paper described WIWO principles of opera-
tion and offers a comprehensive summary of WIWO analytical
characteristics. Simulation and experimental results were also
reported. A modified PWM modulator scheme required to make
the converter work coherently was also suggested. A prototype
WIWO dc–dc converter was built and tested. The converter
demonstrated in practice the WIWO dc–dc conversion ratio.
The new converter topology has several advantages. The
WIWO retains the features of both the buck and the boost con-
verters; however, it achieves wider step-up and wider step-down
dc–dc conversion range. The WIWO converter can operate with
an input source with broadly varying voltage or, alternatively,
feed loads with variable operating voltage such as dc motors.
The converter has a simple structure and moderate component
count. The advantageous buck feature allows turning off the
output voltage on demand. WIWO is also inherently capable of
limiting the inrush current and can protect the output in the case
of a short circuit. Due to the nonlinear characteristics, WIWO
can avoid operating at extreme duty cycle. As a result, WIWO
efficiency remains high even throughout large input voltage
swing. The transition between the operating modes is inher-
ently smooth, and causes no transient disturbance in the average
current. Among the disadvantages of WIWO is the coupled in-
ductor whose leakage causes oscillation and high voltage spike
across the switches. Clamp circuits are needed to clamp volt-
age spikes upon switches, so as to recycle the leakage energy.
Another disadvantage of WIWOs is that small-signal transfer
functions include an RHP zero, and therefore, WIWO is some-
what difficult to stabilize using a single voltage loop. To resolve
the dynamic problem, current loop should be employed, which
is a good practice in any case. An additional disadvantage is that
WIWO does not provide isolation. This, however, may not be
much of a problem in systems with multiple stages.
Modifications of the WIWO to synchronous WIWO dc–dc
converter, bidirectional WIWO dc–dc converter, and WIWO
dc–dc converter for PFC are possible. Numerous advantages
indicate WIWO as a viable candidate for many industrial
applications.
REFERENCES
[1] D. Maksimovic and S. Cuk, “Switching converter with wide dc conversion
range,” IEEE Trans. Power Electron., vol. 6, no. 1, pp. 151–157, Jan. 1991.
[2] K. Yao, M. Ye, M. Xu, and F. C. Lee, “Tapped-inductor buck converter for
high-step-down dc–dc conversion,” IEEE Trans. Power Electron., vol. 20,
no. 4, pp. 775–780, Jul. 2005.
[3] J.-H. Park and B.-H. Cho, “Nonisolation soft-switching buck converter
with tapped-inductor for wide-input extreme step-down applications,”
IEEE Trans. Circuits Syst. I, Reg. Papers, vol. 54, no. 8, pp. 1809–1818,
Aug. 2007.
[4] K. Yao, Y. Ren, J. Wei, M. Xu, and F. Lee, “A family of buck type dc–dc
converters with autotransformers,” in Proc. Appl. Power Electron. Conf.
Expo. (APEC 2003), pp. 114–120.
Authorized licensed use limited to: RAJESH K. Downloaded on March 19,2010 at 01:54:59 EDT from IEEE Xplore. Restrictions apply.
289
[5] K. Nishijima, K. Abe, D. Ishida, T. Nakano, T. Nabeshima, T. Sato, and
K. Harada, “A novel tapped-inductor buck converter for divided power
distribution system,” in Proc. IEEE PESC Conf. (PESC 2006), Jun., 18–
22, pp. 1–6.
[6] G. Spiazzi and S. Buso, “Power factor preregulator based on modified
tapped-inductor buck converter,” in Proc. IEEE PESC Conf., 1998, vol. 2,
pp. 873–879.
[7] F. L. Luo and H. Ye, “Positive output cascade boost converters,” Proc.
Inst. Electr. Eng. Electr. Power Appl., vol. 151, no. 5, pp. 590–606, Sep.
2004.
[8] Q. Zhao and F. C. Lee, “High efficiency, high step-up dc–dc converters,”
IEEE Trans. Power Electron., vol. 18, no. 1, pp. 65–73, Jan. 2003.
[9] N. Vazquez, L. Estrada, C. Hernandez, and E. Rodriguez, “The tapped-
inductor boost converter,” in Proc. IEEE Int. Symp. Ind. Electron., Jun.,
4–7, 2007, pp. 538–543.
[10] K. C. Tseng and T. J. Liang, “Novel high efficiency step-up converter,”
Proc. Inst. Electr. Eng. Electr. Power Appl., vol. 151, no. 2, pp. 182–190,
Mar. 2004.
[11] H. Cheng, “Wide input wide output (WIWO) dc–dc converter,” Master’s
thesis, Univ. California, Irvine, Dec. 2007.
[12] H. Cheng and K. Smedley, “Wide input wide output (WIWO) dc–dc con-
verter,” in Proc. IEEE Appl. Power Electron. Conf. Expo., 2008, pp. 1562–
1568.
[13] R. D. Middleblook, “A continuous model for the tapped-inductor boost
converter,” in IEEE PESC 1976 Rec., Jun., pp. 10–19.
[14] D. Edry, M. Hadar, O. Mor, and S. Ben-Yaakov, “A SPICE compatible
model of tapped-inductor PWM converters,” in Proc. IEEE APEC 1994,
pp. 1021–1027.
[15] R. Erickson and D. Maksimovic, Fundamentals of Power Electronics.
Norwell, MA: Kluwer, 2001.
Hao Cheng, photograph and biography not available at the time of publication.
Keyue Ma Smedley (S’87–M’90–SM’97–F’08) re-
ceived the B.S. and M.S. degrees in electrical engi-
neering from Zhejiang University, Hangzhou, China,
in 1982 and 1985, respectively, and the M.S. and
Ph.D. degrees in electrical engineering from the
California Institute of Technology, Pasadena, in 1987
and 1991, respectively.
She was employed at the Superconducting Super
Collide from 1990 to 1992, where she was respon-
sible for the design and specification of ac-dc con-
verters for all accelerator rings. She is currently a
Professor in the Department of Electrical Engineering and Computer Science,
University of California at the Irvine (UCI). She is also the Director of the UCI
Power Electronics Laboratory. Her research activities include high efficiency
dc-dc converters, high-fidelity class-D power amplifiers, active and passive soft
switching techniques, single-phase and three-phase PFC rectifiers, active power
filters, grid-connected inverters for alternative energy sources, VAR on demand
for modern grid, motor drive, fault current limiters for utility, solar and wind
power conversion, etc. She has authored or coauthored more than 100 technical
articles and holds ten US/international patens.
Dr. Smedley is the recipient of UCI Innovation Award 2005.
Alexander Abramovitz (M’06) was born in
Kishinev, USSR, and repatriated to Israel in 1973.
He received the B.Sc., M.S., and Ph.D. degrees all in
electrical engineering from Ben-Gurion University
in the Negev, Beer-Sheva, Israel, in 1987, 1993, and
1997, respectively, and a postdoc from the University
of California, Irvine, in 2004.
He is currently an Assisting Professor in the De-
partment of Electrical and Electronics Engineering,
Sami Shamoon College of Engineering, Beer-Sheva
and a Visiting Researcher in the Department of Elec-
trical Engineering and Computer Science, University of California. His current
research interests include switch mode and resonant power conversion, high
quality rectification, electronic instrumentation, and analog circuits.