Modeling of a SEPIC Converter Operating in

Discontinuous Conduction Mode

Vuthchhay Eng and Chanin Bunlaksananusorn
Faculty of Engineering, King Mongkut's Institute of Technology Ladkrabang (KMITL),
Chalongkrung Rd. Ladkrabang, Bangkok 10520
vouthchhay_oeung@yahoo.com and kbchanin @kmitl.ac.th
Abstract - A SEPIC (Single-Ended Primary Inductor Converter)
DC-DC converter is capable of operating in either step-up or stepdown mode and widely used in battery-operated equipment. There
are two possible modes of operation in the SEPIC converter:
Continuous Conduction Mode (CCM) and Discontinuous
Conduction Mode (DCM). This paper presents modeling of a SEPIC
converter operating in DCM using the State-Space Averaging (SSA)
technique. The modeling leads to a small-signal linear model of the
converter, from which the transfer functions used for feedback
control design can be determined. It is found that the derived model
is a reduced-order model, which can accurately predict the
converter's characteristics up to one-tenth of the switching
frequency.

I.

INTRODUCTION

A SEPIC (Single Ended Primary Inductor Converter) converter

is a fourth-order dc-dc converter capable of delivering an output
voltage which can be greater or lower than an input voltage.
There are two possible modes of operation in the SEPIC
converter: Continuous Conduction Mode (CCM) and
Discontinuous Conduction Mode (DCM). Although the converter
may have been designed for the CCM operation, it can plunge
into the DCM operation at light loads. In some cases, the
converter is even intentionally designed to operate in DCM
because of the faster dynamic response compared with the CCM
[1, 2]. Recently, the small-signal dynamic characteristics of the
DCM SEPIC converter have been modeled [3]. In this work, the
active switch and passive diode of the SEPIC converter were
substituted by the PWM-switch model [4]. Transfer functions of
interest, e.g. duty ratio-to-output or input-to-output transfer
function, were then derived from the resulting equivalent circuit.
However, the modeling process in [3] had assumed the converter
is being ideal and neglected the Equivalent Series Resistance
(ESR) of the capacitors. The ESR affects the value of zeros in the
final transfer functions - excluding it from the modeling process
only adds to the inaccuracy in the final model.
This papers presents modeling of the DCM SEPIC converter
with State-Space Averaging (SSA) technique [1], taking into
account the effect of the capacitors' ESR. The transfer functions
for feedback control design are derived and compared with the
results from the DCM averaged switch model [2]. It is found that
the derived model is a reduced-order model, which can

978-1-4244-3388-9/09/$25.00 2009 IEEE

accurately predict the converter's characteristics up to one-tenth

of the switching frequency.
II. OVERVIEW OF SSA TECHNIQUE IN DCM
A SEPIC converter is shown in Fig. l(a). Due to the switching
action of the MOSFET, Q, and diode, D, the converter will
exhibit three different circuit states in one switching period, T,
when operating in DCM. The first state exists when Q is turned
on for a time interval d j T (Fig. 1(b)), the second state when Q is
turned off (i.e. D turned on) for a time interval d2 T (Fig. l(c)),
and the third state when both Q and D are turned off for the rest
of the time period d3 T (Fig. l(d)). Note that d j and d2 are the duty
cycle of Q and D respectively, and d3 equals l-d j-d2. The general
state-space equations for these three circuit states are:

dxldt = A1x+B1u
{y=C1x+E1u

~ tIme
. .IntervaIdT
lor
1

(I-a)

dxl dt = A 2 x+B 2 u
{y=C x+E u
2
2

~ tIme
. .IntervaI d2T
lor

(I-b)

dxldt = A3 x+B3u
{ y=C3 x+E3 u

~ tIme
. .IntervaIdT
lor
3

(I-c)

Since the SEPIC converter is made up of two inductors L j and L 2

and two capacitors C j and C2, the state vector x, thus, comprises
of iLj, iL2 , VCj, and VC2. The input voltage vg is typically assigned
as the input vector u, and the output voltage Va as the output
vector y. To find the averaged behavior of the converter over one
switching period, T, equations (I-a) to (I-c) are weighed average
by the duty cycles as:
dXldt=AX+BU
{y=Cx+Eu
(2)
where
A = A1dl+ A 2d2+ A 3d3,
B = B1dl + B 2d2+ B 3d3,
C=Cldl+C2d2+C3d3' and
E=Eldl+E2d2+E3d3.
Equation (2) is a nonlinear continuous-time equation. It can be
linearized by small-signal perturbation with x=X+x, y=Y+y,
u=U+u, dj=Dj+dj, d2=D2+d2, and d3 =D3-dj-d2 where the tilde
symbol " ,." "represents a small-signal value and the capital letter
a DC value. It should be noted that Xx, Yy, Uu, Djdj,
D2d2, and D 3 d3 The perturbation yields the steady-state
and linear small-signal state-space equations in (3) and (4)
respectively.
dXldt=AX+BU=O
(3)
{Y=CX+EU

~i/dt_=A~+Bii~BdlJl+_Bd2J2
{y=Cx+Eu+Edldl+Ed2d2

(4)

where

A=AIDI+~D2+A3D3' B= BIDI+B2D2+B3D3, C=CIDI+C2D2+C3D3'

E=EI~+E2D2+E3D3'
Bdl = (AI-A3)X+(BI-B3)U, Bd2=(A2-A3)X+(B2-B3)U,
Ed. = (C.-C3)X+(E.-E3)U, and Ed2=(C2-C3)X+(E2-E3)U.

Equations (1) to (4) provide a systematic way to model the DCDC converters in DCM. When applied to the DCM SEPIC
converter, the direct solution of (3) and (4) will not produce
correct results because the chosen state variables, iLl and i L2, are
actually not independent from each other. Hence, only one of
these currents can be said to be a true state variable. If both iLl
and i L2 are to remain as a state variable as in equations (2) to (4),
some constraints must be imposed on them. As shown in [1], the
steady-state and linear small-signal state-space equations in (3)
and (4) are subject to the following constraints:
I =i(~,Vo,4,~,T)

(5-a)

where I =ILl + IL2 .

dlldt=O
{1=(alIOvg)vg+(alIOvo)vo+(allad)dl
(5-b)
wehre I =~1+~2
To find the steady-state solution, due to the reason above, the
relationship between ILl and I L2 must be first established before
the other unknown, such as VC1 VC2 , and Vo , can be found by
solving (3):
X = -A -I BU
{ Y = (-CA -I B + E)U

A.

State-Space Description

Chosen x = [iLl i L2 VC1 VC2]T, U = v g , and y = vo, the matrices

At, A2 , A3 , B t , B 2, B3 , C t , C 2 , C 3 , E t , E 2, and E 3 in (1) are

:e:r[t~j:o?:the cgcuitS]in Figs. l(b) to l(d).

.- 0 -I/C 0

(8-a)

0 I 0 -REI(C2rc2R)
(RE- rCI)/4 -REI4 -114 -RE I(4 rC2)]
A - -REIL2
-REIL2
0 -REI(L2rc2 )
2- llC
0
0
0
[
R/(C2rC2 ) R/(C2rC2 ) 0 -REI(C2rc2R)
rC /(4 +~) 0 -1/(4 +L2) 0
]
A - -rc/(4+~) 0 1/(4+ L2) 0
3- l/C
0
0
0
[ I
o
0 0 -REI(C2rc2R)
B.=[1/4 0 0 OJT

III. MODELING OF SEPIC CONVERTER IN DCM

In Fig. l(a), the resistances rC1 and rC2 are Equivalent Series
Resistances (ESRs) of the capacitors C1 and C2 respectively.
Although their values are small, these ESRs cannot be neglected
in the modeling process as they have a direct impact on the
accuracy of the final model. Fig. 2 depicts the current waveforms
iLl and i L2 of the converter, when operating in DCM. The currents
iLl and i L2 are increasing during the time interval d1T and
decreasing during the time interval d2 T. During the time interval
d3T, these currents have a constant value, with the amplitude of
iLl and i L2 being equal but flowing on the opposite direction (Fig.
l(d; that is, iLl = -iL2 The equality in the amplitude of iLl and i L2
is always true during the time interval d3T, and this essentially
makes iLl and i L2 depend on each other.

(8-c)
(8-d)
(8-e)
(8-f)

B 2 =[1/4 0 0 OJ T

B3=[1/(4+L2) -1/(4+L2) 0 OJT

c. =[0

0 0 RE lrc2J

(8-g)

C 2 =[RE RE 0 RElrc2 J
C 3 =[0 0 0 RE lrc2J
E.=E2 =E3 =[0]

(8-h)
(8-i)
(8-j)

(6)

The constraints in (5) will result in 32 and one inductor current be

eliminated from the linear small-signal state-space equations in
(4). The disappearance of one inductor current means the
system's order has been reduced by one. For this reason, the
converter's model derived by the SSA technique in DCM is
known as a reduced-order model [3, 5]. Consequently, the linear
small-signal state-space can be rewritten as:
~i/dt~A...i~BmVg +~mdlJl
(7)
y = Cmx+Em Vg + Emd.dl
Finally, by applying the Laplace transform to (7), transfer
functions for feedback control design, e.g. duty ratio-to-output or
input-to-output transfer function, can be derived.

(8-b)

(a) SEPIC converter.

CI rCI iCI

R
(b)

SEPIC converter during the first state d]T.

iLl

CI rCI
+ -

i CI

(c) SEPIC converter during the second state d2T.

iLl

CI rCI
+ -

i CI

(d) SEPIC converter during the third state d3T.

Fig. 1. Operation of the SEPIC converter in DCM.

Fig. 2. Current waveform of iLl and i12.

Referring back to (3) and (4), the matrices for the steady-state
equations and the linear small-signal state-space equations are
determined from (8):
ret D3 (ret +RE)D2 -RED2 _ K
-RED2
4+ L2
4
4
34
4 rC2
-ReDz _ ret D3 -retDl - RED2 K
-ReDz
A=
L2
4 +L2
L2
34
L2rC2
D2+D3
-Dl
0
0

Cl

D 2=~217LE/(RT)

(9-a)

RE D2
RE D2
0 ~
C2rC2
C2rC2
C2rC2 R
B=[(Dl+D2)/4+D3/(Ll+L2) -D3/(4+ L2) 0 OJ
C=[RED2 RED2 0 RElrc2 J
E=[O]

]
B = rclILl 1(4 +L2)+(~1- rclI L2 )1L2
dl
-(ILl + IL2)/~
[

(9-e)

d2

-[(rCl +RE)I4+ rc/(4 +~)]ILl-(IL2+Vc2Irc2)REI 4]

[rc/(4+L2)-REI4]ILl-(IL2+~2Irc2)REIL2

0
RE(ILI + I L2 )/( C2rC2 )

vo=Cm[ZLl ZL2

Steady-state equations

(11)

In steady-state, I C1 is equal to zero. Thus the relationship between

ILl and I L2 is obtained as:

= I Ll D 2 / D l

VCl VC2 ]

+ Emvg + Emdtdl

where

(12)

Given the averaged matrices in (9-a) to (9-d) and the relationship

in (12), the steady-state solution of converter can be obtained by
using (5).

ILl ]
(~I D2)21]1R
IL2 =v 1]Dl /(RD2)
(13)
V
g
1
[ Cl
VC2
1]Dl l D2
where 1]=1/[1+REDI I(RD2)+[4 +~(I-Dl)1 D2]rClDl/[RD2(4 +~)]]

[ lICI
rC lK12 /(nC2)

(9-f)

As stated above, the relationship between the steady-state

inductor currents, ILl and I L2 , must be determined before other
steady-state values can be found. This is done by averaging the
capacitor current i C1 in Fig. 3 over a switching period, T. In Fig. 3,
i C1 can be expressed as:
i Cl = -iL2
for time interval dlT
(10-a)
i Cl = iLl
for time interval d 2T
(10-b)
i Cl = iLl = -iL2
for time interval d 3T
(10-c)
Average of i C1 over a switching period, T, gives:

I L2

;l~l ~: V~ v~zJ ~A;[~l ~2 vel v~J + BmV + Bmd1dl (16)

rC1(K23-KtJ2KI2In) 0 -KJ2KI2In-TDIKEI(2~)-K34
A = rC1(KJ2K12ln-K23) 0 KJ2KI2In+TDIKEI(2L)+K34

(9-g)
(9-h)

I a =Imin -(IL2+Imin)Dl/(Dl+D2)+(ILl-lmin)D2/(Dl+D2)

Linear Small-Signal State-Space Equations

Referring to (7), linear small-signal state-space equation is:

D2[KJ/(nLE)-1/4]]
-D2[KJ/(nLE)-1/4]

-TDI2/(2~CI)

0 (D2- IRR /

n)TDI/(2~C2)+ IKI/(nC,)

-[IIR+ID2/(nLE)]/C2

(Dl +D2)1 4+ D31(4 +~)- KalDl+D2)/(n4)-TDlKEI(24) ]

= -(Dl+D2)/4-D3/(4+~)+KalDl+D2)/(n4)+TDlKEI(24)

Edl=[O]
Ed2=[RE(ILl+ I L2 )]

B.

(9-b)
(9-c)
(9-d)

(15)

where LE=(4+~)/(4~) and RE=(R+rC2)/(RrC2).

c.

-rClIL/(4+L)+~/4

From Fig. 2, the average value of ILl and I L2 are expressed as:
ILl =DlT(Dl + D 2)Vg1(24) +I min
(14-a)
IL2=DlT(Dl+D2)Vgl(2L2)-Imin
(14-b)
Matching the summation of ILl and I L2 in (13) to those in (14), D 2
can be defined as:

2
[_ TD11(24Cl)
(Dl+D2)Il(n4C2)+(D2- IRR/ n)TD/(24C2)
Cm =[rClREIK12 ln 0 RE[IK12ln+(D2-IRRJn)T~/(2~)] l-IRED2/(nLE)]
m

Em=[REI(~+D2)/(n4)+(D2-IRRrln)T~REI(24)]

mdl

~[11 4- Ka2/(nLE)]+rCl[Ka2IL2I(n~)-ILl (4 +~)]- KTIKEIDl ]

= -Vg[lI 4- Ka2/(nLE)]-rCl[Ka2IL2/(n~)-ILl (4 + ~)]+ KTIKEIDl

-2IICI
(VgI LE-rCl IL2 1~)II(nC2)+(D2- IRR/ n)II(DlC2)

E mdl =[(VgILE-rcl L2 1~)REI In+(D2-IRRJn)REI IDlJ

K12 =Dl/~-D2/4, K 23 =D3/(4 +~)-D2/4, K34 =D2/4 +D3/(4 +L2),

Ka2=REII4+Vc2/4, n=:::.(REI+Vc2 )ILE, 1]=:::.1,
KT=I/[I+TDlrcl/(2~)]=:::.I, I=:::.DlTVgl(2LE), and RRr=rClDl/~+RED2ILE.
KE=RED2/4-Ka2~Rln,

D.

Finding Transfer Function

Applying the Laplace transform to (16), various transfer

functions can be derived for the converter. Due to the limited
space, only two important transfer functions are presented: the
duty ratio-to-output voltage and input-to-output transfer
functions, Gd1v(S) and Gvv(s), respectively.
Gdly(S)=vo(s)1 dl(s)=C(sI - A)-tBdl +Edl
2
[TD2~/(LEC2)](adlvS3 +bdlvS +CdlvS + ddlv)
[as+[2+ RE/(RD2)]/(C2R)](S2 +bs+c)
Gyy(s) = Vo(s)1 vg(s) =C(sI - A)-tBot + E ot
[TD2/(4LiL2ClC2)](aws3 +b s 2+cws+d )
vv

vv

(17)

(18)

[as+[2+ RE/(RD2)]/(C2R)](S2 +bs+c)

Coefficients of Gd1v(S) and Gvv(s) are listed in TABLE I.

Irrnnl:

~_

............,.I~ ......

O~--+--

:d~t: d T id
...

~I..

i
3

I
I

~I" ~I

Fig. 3. Current waveform oficl.

IV. RESULTS
Fig 4(a) shows a frequency response of Gd1v(S) in (17). The
plot is generated by MATLAB using the SEPIC converter's
circuit parameters in TABLE II (these values cause the converter

to operate in OCM). Fig. 4(b) shows the same plot from PSPICE
simulation using the OCM averaged switch model [2]. Generally
accepted as being accurate, this PSPICE result is used to validate
Gd1v(S) in (17). The two results are closely agreed at frequencies
below 10 KHz, i.e. one-tenth of switching frequency, beyond
which the result from (17) starts to diverse from its PSPICE
counterparts. The discrepancy occurs due to the fact that the
former is a reduced-order model, while the latter is a full-order
model [2, 5]. Gd1v(S) in (17) being a reduced-order model is
evident in (16), where all elements in the second column of Am
matrix are zero, which literally means i L2 is no longer a state
variable, thus reducing the system's order by one.
TABLE I
COEFFICIENTS OF Gd1v(S) AND Gvv(s).
ad1v =C2RE[1- TD1rCI 1(4~)- D1LErCI 1(2D2~R)]

bdlv=(l+REIR)[1-TDlrC/(4~)]+[D2/~-D/~+TDI2rc/(4l3i)]TDIC2R E I(2C1 )
D1LEC2RErJI [~b-DzLz [(TDzR +l)D +D]+ DzL,i 2Dz- 1) (TDzR 1)]
2Di~~R
~+~
2LE
2
I
L2
2LE +
-[[TD?~/(4D2~CIR)+1]C2/~+DI(R+RE)/(2D2R2RE)-2C2(DI+D2)/~]LERErCI/~
_ T~ ( RE D2 ~ rp(R+RE) T 2Iif
1
TIif4
cdlv- 2q l+jf)(Z;--Z:;)+
R
[g4C + 4+~ (l-2D3 4D2~qR)]
1

~4C2RErQ +~[l [T D L 4D (T n L D )] T~rp]

2D2CIR(4+~)2 q(4+~) - L.f+ 2~- 3 L.f~-~ 2 4~4
+ ~rp(~-D2) [C2RE (D ,-_'}nT)-!Q.[Tn2c RR +2'-C(R+R )]]
2D2qR(4+~)2
D2
2~ ~LJI 2R ~ 2 E ~ I
E

v.

In this paper, modeling of a OCM SEPIC converter with StateSpace Averaging (SSA) technique has been presented. The
modeling yielded a steady-state and linear small-signal equations
of the converter in (13) and (16) respectively. From (16), the
transfer functions Gd1 v(s) and Gvv(s), which provide a basis for
feedback control design, were derived. The obtained Gd1v(S) was
compared against the result from the OCM averaged switch
model [2]. Good consistency between the two results was
observed only at frequencies below 10kHz, or one tenth of the
switching frequency. Above this frequency, the derived Gd1v(S)
became invalid. Oue to this limitation, if the derived models in
(17) and (18) were to be used in feedback control design, a
crossover frequency must be selected in the frequency region
where these models are legitimate.
REFERENCES
[l]

[2]
[3]

ftr~l_ {+(2~CI-T~2C2RE)(D2~+~4+24D2-4) ~:4;R+RE)

[4]

LF:Dz(2fz-12+A)[~~f (~24C2+D24q+D2~q)+2R+2RE]

[5]

LJI+~ g~4q
4D2~R

2D2R(4+~)

'-1L.f~

_...L..[ qCz4(A-l) +D R
44 R Di~CI(4+~)

]}
E

CONCLUSION

S. Cuk and R. D. Middlebrook, "A general unified approach to modeling

switching DC-to-DC converters in discontinuous conduction mode," in
Proc. IEEE PESC'77, 1977.
R. W. Erickson and D. Maksimovic, Fundamentals of Power Electronics,
2nd ed., Kluwer Academic Publishers, 2001.
L. G. De Vicuna, F. Guinjoan, J. Majo, and L. Martinez, "Discontinuous
conduction mode in the SEPIC converter," Proc. of Electrotechnical
Conference on Integrating Research, Industry and Education in Energy and
Communication Engineering, page. 38-42, April 1989.
E. Niculescu, M. C. Niculescu, and D. M. Purcaru, "Modelling the PWM
SEPIC converter in discontinuous conduction mode," Proc. of the 11 th
WSEAS International Conference on Circuits, July 2007.
Jian Sun, D. M. Mitchell, M. F. Greuel, P. T. Krein, and R. M. Bass,
"Averaged modeling of PWM converters operating in discontinuous
conduction mode," IEEE trans. on Power Electronics, July 2001.

2
d = (l+R,:;IR) {1- TD1rCI (T +D T ) I D1rCI [TDI rCl (T -D,T -D T
dlv CI(~+~)
44~ ~
2~
4D l3iR 2~ ~ l~
2~
2
2
- 2D2~)(2LE- TD2R)+[TDtrCI +2~(DID2 - Dl - D2)]LEID2]
+ DlrClLE(D2~ - DI~) [T (2D T _ 2D T +TD 3,., )-4TD2D R( T + T)] }
4l3i4DiR
~
l~
2~
I CI
2 3 ~ ~
aw = 4 CIC2RE[2L/2D1+D2)-

4 TrCl]

,-..

2
bw ={24~~(l+ REIR)(2D1+ D2)+TD1D2C2RE(4 +~)
+(2D24 - 4 + D14 + D2~)TD12CIC2REr6 1(4 +~)
2
+ 4~rCl[[D14 + 2~(2Dl + D2-1)(D1+ D2)]2C2RE1(4 + ~)-(R + RE)TD1I R] }

cw=(1+REIR)[TD12D2(~+L2)+[2(Dl+D2)(2Dl+D2-1)~+Dl~]2~CrCl(~+L2)
-(2D2~+Dl~-~+D2~)TD12Cr~/(~+~)]
+4Dl~C2RE[[D14-4-D2~-2D2~D3]TD{Cl/(4~~)+1]

dw=(1+REIR)[4Dl~+[Dl~-~-D2~-2D2L2D3]TD12rCl/~]

TABLE II
CONVERTER PARAMETERS

Circuit Parameters

Values

Vg / Vo

15/5V
47 /200JlF
0.38/0.095il
1Oil
100/30flH

C j /C2

rCj/rC2

R
L j /L 2

00
Q)

-45

-90
10~1----'----'--''''''''''''''''....i..i.l.L.:02----i..---i....-i.....i....i...i..i..il..L..:03~----i......i.....i....i...i..i....i.L-----i.---..L...L..Ll...i..i...LJl05

Q)
cr.J
~

Frequency (Hz)
(a)