Dr.Arkan A.

Hussein Power Electronics Fourth Class

‹
C uK and Sepic Converter

١
Dr.Arkan A.Hussein Power Electronics Fourth Class

Instructional objective
On completion the student will be able to

• Compare the advantages and disadvantages of CuK and Sepic converters with those of
three basic converters.

• Draw the circuit diagrams and identify the operating modes of CuK and Sepic
converters.

• Draw the waveforms of the circuit variables associated with CuK and Sepic converters.

• Calculate the capacitor voltage ripples and inductor current ripples in CuK converter.

٢
Dr.Arkan A.Hussein Power Electronics Fourth Class

24.1 Introduction
Switch Mode Power Supply topologies follow a set of rules. A very large number of converters
have been proposed, which however can be seen to be minor variations of a group of basic DC-
DC converters – built on a set of rules. Many consider the basic group to consist of the three:
BUCK, BOOST and BUCK-BOOST converters. The CUK, essentially a BOOST-BUCK
converter, may not be considered as basic converter along with its variations: the SEPIC and the
zeta converters.

The Canonical Cell forms the basis of analyzing switching circuits, but the energy
transport mechanism forms the foundation of the building blocks of such converters. The Buck
converter may consequently be seen as a Voltage to Current converter, the Boost as a Current to
Voltage converter, the Buck-Boost as a Voltage-Current-Voltage and the CUK as a Current-
Voltage-Current converter. All other switching converter MUST fall into one of these
configurations if it does not increase the switching stages further for example into a V-I-V-I
converter which is difficult to realize through a single controlled switch. It does not require an
explanation that a current source must be made to deliver its energy into a voltage sink and vice-
versa. A voltage source cannot discharge into a voltage sink and neither can a current source
discharge into a current sink. The first would cause current stresses while the latter results in
voltage surges. This rule is analogous to the energy exchange between a source of Potential
Energy (Voltage of a Capacitor) and a sink of Kinetic Energy (Current in an Inductor) and vice-
versa. Both can however discharge into a dissipative load, without causing any voltage or current
amplification. The resonant converters also have to agree to some of these basic rules.

24.2 Analysis of C uK converter

The advantages and disadvantages of three basic non-isolated converters can be summerised as
given below.

(i) Buck converter

1
L
2 S1
iB
Vin
C

Fig. 24.1: Circuit schematic of a buck converter

• Pulsed input current, requires input filter.

Features of a buck converter are

• Continuous output current results in lower output voltage ripple.

• Output voltage is always less than input voltage.

٣
Dr.Arkan A.Hussein Power Electronics Fourth Class

(ii) Boost converter

L
1
S1
2
R
Vin
C

Fig. 24.2: Circuit schematic of a boost converter

• Continuous input current, eliminates input filter.

Features of a boost converter are

• Pulsed output current increases output voltage ripple.

• Output voltage is always greater than input voltage.

(iii) Buck - Boost converter

1 2
S1

L R
Vin
C

Fig. 24.3: Circuit schematic of a buck boost converter

• Pulsed input current, requires input filter.

Features of a buck - boost converter are

• Pulsed output current increases output voltage ripple

• Output voltage can be either greater or smaller than input voltage.
It will be desirable to combine the advantages of these basic converters into one converter.

• Continuous input current.

CuK converter is one such converter. It has the following advantages.

• Continuous output current.

• Output voltage can be either greater or less than input voltage.

CuK converter is actually the cascade combination of a boost and a buck converter.
L1 2 L2

S1
1 + +
R
Vin C - C2 -
2'
S2
1'
Fig. 24.4: Circuit schematic of a boost-buck converter

٤
Dr.Arkan A.Hussein Power Electronics Fourth Class

S1 and S2 operate synchronously with same duty ratio. Therefore there are only two switching
states.

(i) 0 < t ≤ DT
S1 to (1)

& S2 to (1')

The circuit configuration is given below

L1 L2 L1 C1

R R
Vin C1 C2 C2

(a)

(ii) DT < t < T; S1 to (2) & S2 to (2')

L1 L1 L2

R R
Vin C2 C2

(b)
Fig. 24.5: Circuit topology of a boost-buck converter during different
switching intervals
(a) 0 ≤ t < DT, (b) DT ≤ t < T
These two topologies can also be obtained from the following circuit which is the so called
CuK converter.

٥
Dr.Arkan A.Hussein Power Electronics Fourth Class

L1 C1 L2

1 2
S
R
Vin C2

(a)
vc1
L1 C1 L2
+ -
ic2 i0
iL1 iL2
1 2 +
Vin R V0
iB C2 -

(b)
Fig. 24.6: Schematic and Circuit representation of ĈuK converter.
(a) Schematic diagram, (b) Circuit diagram

24.2.1 Expression for average output voltage and inductor

currents
L1 L2
+
ic1 ic2 i0
iL1 iL2
- +
R V0
Vin C2 -
C1 +
-
(a) 0 < t ≤ DT
L1 L2
+
ic1 iL2 ic2 i0
iL1
+ +
VC1 R V0
Vin C2 -
C1 - L2
-
(b) DT < t ≤ T
Fig. 24.7: Equivalent Circuit of a ĈuK converter during different
conduction modes.
(a) 0 < t ≤ DT (b) DT < t ≤ T

٦
Dr.Arkan A.Hussein Power Electronics Fourth Class

Applying Volt-sec balance across L1

Vin DT + (Vin − VC 1 ) (1 − D ) T = 0 (24.1)

∴ Vin (1 − D ) VC 1 = 0

VC1 =
1− D
Vin
or (24.2)

Applying Volt-sec balance across L2

(V0 + VC1 ) DT + V0 (1 − D ) T = 0 (24.3)

or V0 + DVC1 = 0 (24.4)

V0 = − DVC1 = −
1− D
DVin
or (24.5)

Expression for average inductor current can be obtained from charge balance of C2

I L 2 + I0 = 0 (24.6)

∴ I L 2 = − I0 = − = D in
R 1− D R
V0 V
(24.7)

From power balance

= V0 I 0 = =
R (1 − D )2 R
v02 D 2 Vin2
Vin I L1 (24.8)

∴ I L1 =
(1 − D ) R
D 2 Vin (24.9)
2

24.2.2 Current ripple and voltage ripple calculations

The waveforms of different circuit variables of Fig. 24.7 are given in Fig. 24.8.

٧
Dr.Arkan A.Hussein Power Electronics Fourth Class

iB

DT T

IL1 MAX
IL1
iL1
IL1 MIN
t
IL2 MAX
IL2
iL2
IL2 MIN

t1 t
t2

ic1 IL2 MAX

IL2 MIN

t
- IL1 MIN
- IL1 MAX
vc1 VC1 MAX
VC1
VC1 MIN
t

ic2

1/2 Iˆ L2 p-p

t1 t2 t

-1/2 Iˆ L2 p-p
Vc2

Vc2

Fig. 24.8: Waveforms of circuit variables in a ĈuK converter.

٨
Dr.Arkan A.Hussein Power Electronics Fourth Class

From the waveforms of Fig. 24.8

I L1MAX = I L1 MIN +
DVin T
(24.10)
L1

= I L1MAX − I L1MIN =
Vin DT
IˆL1 (24.11)
p− p L1

From equation 24.9

I L1MAX + I L1 MIN = 2 I L1 = 2 D 2 Vin

(1 − D ) 2 R
(24.12)

⎡ ⎤ DVin
∴ I L1MAX = ⎢ D 2 + RT ⎥
⎣ (1 − D ) 2 L1 ⎦ R
(24.13)

⎡ ⎤ DVin
I L1MIN = ⎢ D 2 − RT ⎥
⎣ (1 − D ) 2 L1 ⎦ R
(24.14)

I L 2 MAX = I L 2 MIN − (1 − D )T = I L 2 MIN + in DT

V0 V
(24.15)
L2 L2

∴ = I L 2 MAX − I L 2 MIN =
Vin DT
IˆL 2 (24.16)
p− p L2

From equation 24.7

I L 2 MAX + I L 2 MIN = −2 I 0 = 2 D in
1− D R
V
(24.17)

⎡ ⎤ DVin
∴ I L 2 MAX = ⎢ 1 + RT ⎥
⎣1 − D 2 L2 ⎦ R
(24.18)

⎡ ⎤ DVin
I L 2 MIN = ⎢ 1 − RT ⎥
⎣1 − D 2 L2 ⎦ R
(24.19)

For calculating voltage ripples it is noted that

vc1 = 1 ∫ ic1 dt
DT
(24.20)
c1 0

c1 ∫0 c1 c1 ∫0 L 2
but for 0 < t ≤ DT ic1 = iL2 (24.21)
1 DT i dt = 1 DT i dt (24.22)

٩
Dr.Arkan A.Hussein Power Electronics Fourth Class

⎡I + I L 2 MIN RT ⎤ DTI L 2 I 0 DT
vˆc1 = DT ⎢ L 2 MAX + = =
C1 ⎣ 2 L2 ⎥⎦
or (24.23)
2 c1 c1

vˆc1 =
D 2Vin T
RC1(1 − D )
or (24.24)

vˆc 2 = 1 ∫ ic 2 dt which is the hatched area under ic2 waveform in Fig. 24.8
t2
c 2 t1

∴ vˆc = 1 × 1 × T × in =
V DT Vin DT 2
(24.25)
c1 2 2 2 L2 8 L2 C2

Equations 24.11, 24.16, 24.24 and 24.25 can be utilized to design a CuK converter of given
specification

١٠
Dr.Arkan A.Hussein Power Electronics Fourth Class

The SEPIC Converter

The previous chapter discussed the single stage conversion Buck and Boost converters
along with the two-stage Buck-Boost converter. This chapter offers a few additional topologies.

Fig. 24.9(a): A basic converter: BUCK converter

Fig. 24.9(a) is that of a basic Buck converter. From the voltage source C1, the converter charges
the current sink constituted by the inductor-diode (L-D). The current is further converted into
voltage without a switching stage (amplification) at C2. The canonical switching cell is
approached if the capacitors C1 and C2 are combined to be represented by a single capacitor C.
The cell includes T-C-L-D, the basic building block of DC-DC converters. The Boost converter
is realized if the positions of D and T are interchanged in Fig.24.9 (a). Now power flows in from
the right. Here, the energy stored in the inductor during each ON period of switch T is
transferred to the Capacitor during its OFF period.

The CUK converter as the dual of the Buck-Boost converter has current input and current
output stages. The basic SEPIC is a modification of the basic Boost and the CuK topologies.
Consider the Boost converter in Fig 24.9(b). At steady state, the average voltage across the input
inductor is zero. Equating the inductor voltages for the period when the switch T is ON with that

Vin .TON = ( Vout − Vin ) .TOFF

when it is OFF,

or, Vout = ( 1
1− ∂
(24.26)
).Vin
where, ∂ is the duty ratio of the switch.

Fig. 24.9(b): BOOST converter

Fig. 24.10 Modified Boost with load across Diode for Boost-Buck
Operation. (left) without output filter, (right) with filter.

In the path, Vin-L-D-Vout, in Fig. 24.9(b), the average voltages across all the elements are
known. Thus, that appearing across the diode D is Vout – Vin. This voltage from Eqn 1 is:

١١
Dr.Arkan A.Hussein Power Electronics Fourth Class

V D = [(1 1 − ∂ ) − 1].Vin
= (∂
1− ∂
)Vin

A Boost-Buck converter is thus realized. This is the voltage that would appear in an
unfiltered form at the load in Fig. 24.10 (left). Now, since the source is a current source, the
output stage must be capacitive (voltage sink) which is taken care of by C2-D. The voltage across
D has high ripples, which can be filtered much like the Buck converter with an L (and a C3). The
CUK converter is thus realized. It is a I-V-I converter.
A glaring drawback of this derived converter topology is that the polarity of the output is
reversed. This is not acceptable for various reasons.
Now it is the turn of the Diode to be interchanged with the filter inductor. The inductor is
thus converted to be part of the switching circuit and it not just a filter. The SEPIC results – not
an entirely different one - but easily derivable from the previous topologies.

The SEPIC officially stands for “Single-Ended Primary Inductance Converter”. However,
the unofficial interpretation is more descriptive: “Secondary Polarity Inverted Cuk”.

Fig. 24.11(a): The basic SEPIC topology

Again, the basic input–output relation can be derived by considering the two inductors to
have average null voltage across themselves.

If the link capacitor has a voltage Vc across itself (consider it to be reasonably constant),

= (VC − Vout − Vin )TOFF

then for the input inductor, the volt-secs during the ON and OFF periods of the switch are:
Vin .TON
VC = Vout − Vin (. 1
(24.27)
or , )
TOFF

VC .TON = V out .TOFF

For the output inductor,
(24.28)

Eliminating, Vc and writing TON = ∂ . T,

Vout = ( ∂
1− ∂
)Vin (24.29)

Thus the SEPIC is also basically a BOOST-BUCK converter akin to the CUK converter.
(The Boost stage comes first followed by the Buck stage and it is also I-V-I converter)

In the practical SEPIC converter, the two inductors are coupled with the polarities as
indicated by dots in Fig. 24.11(a). The turns ratio is and the coupling is very tight. For such a

(Vin .K .VC ).TON = (Vout + VC − Vin − K .Vout ).TOFF

coupled-transformer SEPIC, equating the positive and negative volt-secs for the two inductors,
(24.30)
for the input inductor, and

١٢
Dr.Arkan A.Hussein Power Electronics Fourth Class

(VC − K ' .Vin ).TON = [Vout − K ' (Vout + VC − Vin )].TOFF (24.31)

Equations (24.28) and (24.29) can be obtained from the above two by substituting both K and K’
to zero to have no coupling between the two coils.

Fig. 24.11(b) The practical SEPIC topology with coupled inductors

The above two equations result in an identity to indicate that such a system cannot work.

This can be explained by examining the operation of the circuit. Initially when the
transistor is OFF, the capacitor C2 charges to the supply voltage Vin. When the transistor is
switched ON, the resulting active circuit is shown in Fig 24.12.

Fig. 24.12: Active part of the circuit when transistor is switched

with C2 charged toVin

The circuits to the left and right of the transistor are identical and both the windings are induced
with the supply voltages, resulting in null emfs on either side, which explains why the ideal
circuit will not work. However, neither the coupling between the inductors nor the effective turns
ratio can be unity. This results in a circuit with the features of the uncoupled circuit and the
circuit performs.

The second voltage source, VC, induces N.VC into the primary, where N is the turns ratio. For the
interesting case, Vin = VC = V1, if the turns ratio, n, is increased slightly from unity, by 1/k (where
k < 1 is the coupling coefficient between windings), then the voltage induced by Vin will increase
the voltage at the Drain of the transistor to N. V1, thereby "bootstrapping" the leakage inductance
of the input inductor. Because the voltage at each end of this leakage inductance is the same, its
inductance is effectively infinite. Consequently, all variations in magnetizing current, (through
M) due to a varying V1 is supplied from the secondary winding source. By symmetry, setting
n = k causes the secondary-winding current to become constant while the primary source
supplies the magnetizing-current variations.

This effect can be desirable because, for n = 1/k, it results in constant (DC) primary current.
Noisy switching current does not appear at the converter input but is diverted instead to the
secondary winding. However, typical values of k are slightly less than one, and turns ratios of
nearly 1:1 may not be easy to wind. One simplification is to use a 1:1 transformer, such as a low-
cost, commodity, common-mode power-line input-filter choke, and add a small additional
inductance in series with the primary winding. This effectively increases the leakage inductance
so that the same secondary-winding dominance of magnetizing current is obtained with n = 1.

١٣
Dr.Arkan A.Hussein Power Electronics Fourth Class

The circuit is an alternative to the Boost converter and outputs an range which includes
the input range also being a Boost-Buck converter. It is superior to the other converters both in
terms of the input current purity and efficiency.

Fig. 24.13: Drain voltages of FLYBACK and SEPIC converters

The waveforms in Fig. 24.13 show the voltage at the transistor Drain present on the fly
back (Boost) and SEPIC circuits. The fly back transformer leakage inductance produces a
voltage spike that adds an additional level to the "flat-top" voltage. This level is about 1.5 times
the supply voltage for inputs around 20 V. In comparison, the SEPIC FET switching waveform
is clamped, and shows very little overshoot, or ringing. This clamping results in less switching-
loss, output voltage noise and a power stage that can be operated at a much higher frequency
than that of the fly back.

Again, the fly back transformer leakage inductance also produces a significant voltage
spike relative to the SEPIC at the output diode. A relatively high voltage (~200V) output diode is
required for the fly back to handle the large negative ringing compared to the SEPIC’s 60V
Schottky diode. The 0.5 volt forward drop of the SEPIC’s Schottky diode relative to the one volt
forward drop of the flyback's ultra-fast diode, results in significant power savings for the SEPIC.

١٤