Chapter 5. The Discontinuous Conduction Mode

Chapter 5.
The Discontinuous Conduction Mode
5.1. Origin of the discontinuous conduction mode, and mode boundary 5.2. Analysis of the conversion ratio M(D,K) 5.3. Boost converter example 5.4. Summary of results and key points
Fundamentals of Power Electronics
Chapter 5: Discontinuous conduction mode
Introduction to Discontinuous Conduction Mode (DCM)

q
Occurs because switching ripple in inductor current or capacitor voltage causes polarity of applied switch current or voltage to reverse, such that the current- or voltage-unidirectional assumptions made in realizing the switch are violated. Commonly occurs in dc-dc converters and rectifiers, having singlequadrant switches. May also occur in converters having two-quadrant switches. Typical example: dc-dc converter operating at light load (small load current). Sometimes, dc-dc converters and rectifiers are purposely designed to operate in DCM at all loads. Properties of converters change radically when DCM is entered: M becomes load-dependent Output impedance is increased Dynamics are altered Control of output voltage may be lost when load is removed
2
5.1. Origin of the discontinuous conduction mode, and mode boundary

Buck converter example, with single-quadrant switches
Q1 iL(t) Vg + D1 iD(t) C R L + V
continuous conduction mode (CCM)

iL(t) I iL
Minimum diode current is (I iL) Dc component I = V/R Current ripple is (V V) Vg DD'Ts iL = g DTs = 2L 2L Note that I depends on load, but iL does not.
0
conducting devices: Q1
DTs
D1
Ts
Q1
iD(t) I iL
DTs
Ts
Reduction of load current

Increase R, until I = iL
CCM-DCM boundary
iL(t)
I 0 DTs
Q1 D1
iL Ts
Q1
Minimum diode current is (I iL) Dc component I = V/R Current ripple is (V V) Vg DD'Ts iL = g DTs = 2L 2L Note that I depends on load, but iL does not.
4
conducting devices:
iD(t)
I 0 DTs Ts
iL t
Further reduce load current

Increase R some more, such that I < iL
I 0
conducting devices: D1Ts Q1
Discontinuous conduction mode

iL(t)
Minimum diode current is (I iL) Dc component I = V/R Current ripple is (V V) Vg DD'Ts iL = g DTs = 2L 2L Note that I depends on load, but iL does not. The load current continues to be positive and non-zero.
5
DTs
D2Ts D1 D3Ts
Ts X
Q1
iD(t)
DTs
D2Ts
Ts
Mode boundary
I > iL for CCM I < iL for DCM
Insert buck converter expressions for I and iL : DVg DD'TsVg < R 2L Simplify:
2L < D' RTs
This expression is of the form
where
K < K crit(D) for DCM K = 2L and K crit(D) = D' RTs
K and Kcrit vs. D

for K < 1: for K > 1:
K < Kcrit: DCM

Kc ( rit D) = 1 D
K > Kcrit: CCM
K > Kcrit: CCM K = 2L/RTs
K = 2L/RTs
0
Kc ( rit D) = 1D
0 0 1
Critical load resistance Rcrit
Solve Kcrit equation for load resistance R:
where
R < Rcrit(D) for CCM for DCM R > Rcrit(D) Rcrit(D) = 2L D'Ts
Summary: mode boundary

K > K crit(D) K < K crit(D) or or R < Rcrit(D) R > Rcrit(D) for CCM for DCM
Table 5.1. CCM-DCM mode boundaries for the buck, boost, and buck-boost converters Converter Buck Boost Buck-boost K crit(D) (1 D) D (1 D)2 (1 D)2
0D1
max ( K crit )
1 4 27 1
R crit(D) 2L (1 D)T s 2L D (1 D) 2 T s
0D1
min ( Rcrit ) 2 L Ts 27 L 2 Ts
2L (1 D) 2 T s
2 L Ts
5.2.
Analysis of the conversion ratio M(D,K)
Analysis techniques for the discontinuous conduction mode:

Inductor volt-second balance
vL = 1 Ts
Ts
vL(t) dt = 0
0
Capacitor charge balance
iC
= 1 Ts
Ts
iC(t) dt = 0
0
Small ripple approximation sometimes applies:

v(t) V because v << V i(t) I is a poor approximation when i > I
Converter steady-state equations obtained via charge balance on each capacitor and volt-second balance on each inductor. Use care in applying small ripple approximation.
10
Example: Analysis of DCM buck converter M(D,K)

iL(t) L + vL(t) iC(t) R + v(t)
subinterval 1
Vg
iL(t)
L iC(t) R + v(t) L
subinterval 2
Vg +
+ vL(t) C
iL(t)
+ vL(t)
iC(t) R
+ v(t)
subinterval 3
Vg
11
Subinterval 1
iL(t)
L + vL(t) iC(t) R + v(t)
vL(t) = Vg v(t) iC(t) = iL(t) v(t) / R

Vg +
Small ripple approximation for v(t) (but not for i(t)!):
vL(t) Vg V iC(t) iL(t) V / R
12
Subinterval 2
vL(t) = v(t) iC(t) = iL(t) v(t) / R

Vg
iL(t)
L iC(t) R + v(t)
+ vL(t) + C
Small ripple approximation for v(t) but not for i(t):
vL(t) V iC(t) iL(t) V / R
13
Subinterval 3
vL = 0, iL = 0 iC(t) = iL(t) v(t) / R

Vg +
iL(t)
L + vL(t) C iC(t) R + v(t)
Small ripple approximation:
vL(t) = 0 iC(t) = V / R
14

vL(t) Vg V
D1Ts D2Ts D3Ts
0 Ts V t
Volt-second balance:
vL(t) = D1(Vg V) + D2( V) + D3(0) = 0

Solve for V:
V = Vg
D1 D1 + D2
note that D2 is unknown
15

node equation:
L iL(t) v(t)/R iC(t) C R + v(t)
iL(t)
Vg V L
iL(t) = iC(t) + V / R
capacitor charge balance:
iC = 0
hence
iL = V / R
must compute dc component of inductor current and equate to load current (for this buck converter example)
ipk
V L
<iL> = I
0
D1Ts
DTs
D2Ts D3Ts
Ts
16
Inductor current waveform

peak current:
iL(t)
Vg V L
Vg V iL(D1Ts) = i pk = D 1T s L
average current:
<iL> = I
ipk
V L
iL = 1 Ts
Ts
iL(t) dt
0
0
D1Ts
DTs
D 2T s D3 Ts
Ts
triangle area formula:

Ts 0
iL(t) dt = 1 i pk (D1 + D2)Ts 2

D 1T s (D1 + D2) 2L
equate dc component to dc load current:
iL = (Vg V)
V = D1Ts (D + D ) (V V) 1 2 g R 2L
17
Solution for V
Two equations and two unknowns (V and D2):
V = Vg
D1 D1 + D2
(from inductor volt-second balance)
V = D1Ts (D + D ) (V V) 1 2 g R 2L
Eliminate D2 , solve for V :
(from capacitor charge balance)
V = 2 Vg 1 + 1 + 4K / D 2 1 where K = 2L / RTs valid for K < K crit
18
Buck converter M(D,K)

1.0 K = 0.01 0.8 K = 0.1 0.6
M(D,K)
0.4
K = 0.5
D M= 1+ 2 1 + 4K / D 2
for K > K crit for K < K crit
0.2
K1
0.0 0.0 0.2 0.4 0.6 0.8 1.0
19
5.3. Boost converter example

i(t) L + vL(t) Vg + Q1 C D1 i (t) D iC(t) R + v(t)
Mode boundary:
Previous CCM soln:
I > iL for CCM I < iL for DCM
I=
Vg D' 2 R
iL =
Vg DTs 2L
20
Mode boundary
0.15
Vg DTsVg > 2L D' 2R
4 Kcrit ( 1 ) = 27 3
for CCM
Kcrit(D)
2L > DD' 2 RTs
for CCM
0.1
where
K > K crit(D) for CCM for DCM K < K crit(D) K = 2L and K crit(D) = DD' 2 RTs
0.05
0 0 0.2 0.4 0.6 0.8 1
21
Mode boundary
0.15
CCM
DCM K < Kcrit
CCM K > Kcrit K
where
K > K crit(D) for CCM for DCM K < K crit(D) K = 2L and K crit(D) = DD' 2 RTs
0.1
0.05
(D) K crit
0 0 0.2 0.4 0.6 0.8 1
D
22
Conversion ratio: DCM boost

i(t) L + vL(t) iC(t) C R + v(t)
subinterval 1
D1 i (t) D iC(t) Q1 C R +
Vg
i(t)
L + vL(t)
i(t)
L + vL(t) iC(t) C R + v(t)
subinterval 2
v(t)
Vg
Vg
i(t)
subinterval 3
Vg
23
Subinterval 1
i(t)
vL(t) = Vg iC(t) = v(t) / R

Vg
Small ripple approximation for v(t) (but not for i(t)!):

vL(t) Vg iC(t) V / R
0 < t < D1Ts
24
Subinterval 2
i(t)
vL(t) = Vg v(t) iC(t) = i(t) v(t) / R

Vg +
Small ripple approximation for v(t) but not for i(t):

vL(t) Vg V iC(t) i(t) V / R
D1Ts < t < (D1 +D2)Ts
25
Subinterval 3
i(t)
vL = 0, i = 0 iC(t) = v(t) / R
Vg +
Small ripple approximation:

vL(t) = 0 iC(t) = V / R
(D1 +D2)Ts < t < Ts
26

vL(t) Vg
D1Ts D2Ts D3Ts
0 Ts Vg V t
Volt-second balance:
D1Vg + D2(Vg V) + D3(0) = 0
Solve for V: D + D2 V= 1 Vg D2
note that D2 is unknown
27

node equation:
iD(t) = iC(t) + v(t) / R
D1 i (t) D iC(t) C R + v(t)
capacitor charge balance:
iC = 0
hence
iD = V / R
must compute dc component of diode current and equate to load current (for this boost converter example)
28
Inductor and diode current waveforms

i(t)
peak current: Vg i pk = DT L 1 s average diode current:
Vg L
ipk
Vg V L
iD
= 1 Ts
Ts
iD(t) dt
0
DTs
D 1T s D2Ts D3Ts
Ts
triangle area formula:

Ts 0
iD(t)
iD(t) dt = 1 i pk D2Ts 2
ipk
Vg V L
<iD> 0
D1Ts
DTs
D2Ts D3Ts
29
Ts
Equate diode current to load current
average diode current:

iD 1 1 i D T = V g D 1 D 2T s = Ts 2 pk 2 s 2L
equate to dc load current:

V g D 1 D 2T s V = R 2L
30
Solution for V
Two equations and two unknowns (V and D2): D + D2 (from inductor volt-second balance) V= 1 Vg D2
V g D 1 D 2T s V = R 2L
(from capacitor charge balance)
Eliminate D2 , solve for V. From volt-sec balance eqn: Vg D2 = D1 V Vg Substitute into charge balance eqn, rearrange terms:
V 2D 2 g 1 V VVg =0 K
2
31
Solution for V
V 2D 2 g 1 V VVg =0 K
2
Use quadratic formula:
V = 1 Vg
1 + 4D 2 / K 1 2
Note that one root leads to positive V, while other leads to negative V. Select positive root:
V = M(D ,K) = 1 + 1 Vg
where valid for
1 + 4D 2 / K 1 2
K = 2L / RTs K < Kcrit(D)
Transistor duty cycle D = interval 1 duty cycle D1

32
Boost converter characteristics

5 M(D,K) 4
K=
0.0 1
K
2
5 0.0
M =
1 1D 1+ 1 + 4D 2 / K 2
for K > K crit for K < K crit
0 K=
.1
K
1
4/2
Approximate M in DCM:
0 0 0.25 0.5 D 0.75 1
M1+ D 2 K
33
Summary of DCM characteristics
Table 5.2. S ummary of CCM-DCM characteristics for the buck, boost, and buck-boost converters Converter Buck Boost Buck-boost with K crit (D) (1 D) D (1 D)2 (1 D)2 DCM M(D,K) 2 1 + 1 + 4K / D 2 1 + 1 + 4D 2 / K 2 D K DCM D2(D,K) K M(D,K) D K M(D,K) D
K
CCM M(D) D 1 1D D 1D
K = 2L / RT s.
DCM occurs for K < K crit .
34
Summary of DCM characteristics
DCM M(D,K)
o Bo
st
c Bu
1
o k-b
os
t(
1)
1 K
DCM buck and boost characteristics are asymptotic to M = 1 and to the DCM buck-boost characteristic DCM buck-boost characteristic is linear CCM and DCM characteristics intersect at mode boundary. Actual M follows characteristic having larger magnitude
Buck
0 0 0.2 0.4 0.6 0.8 1
D
35
DCM boost characteristic is nearly linear

Summary of key points

1. The discontinuous conduction mode occurs in converters containing current- or voltage-unidirectional switches, when the inductor current or capacitor voltage ripple is large enough to cause the switch current or voltage to reverse polarity. 2. Conditions for operation in the discontinuous conduction mode can be found by determining when the inductor current or capacitor voltage ripples and dc components cause the switch on-state current or off-state voltage to reverse polarity. 3. The dc conversion ratio M of converters operating in the discontinuous conduction mode can be found by application of the principles of inductor volt-second and capacitor charge balance.
36
Summary of key points

4. Extra care is required when applying the small-ripple approximation. Some waveforms, such as the output voltage, should have small ripple which can be neglected. Other waveforms, such as one or more inductor currents, may have large ripple that cannot be ignored. 5. The characteristics of a converter changes significantly when the converter enters DCM. The output voltage becomes loaddependent, resulting in an increase in the converter output impedance.
37

Chapter 5. The Discontinuous Conduction Mode

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 5. The Discontinuous Conduction Mode

Uploaded by

Copyright:

Available Formats

Chapter 5.

The Discontinuous Conduction Mode

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Introduction to Discontinuous Conduction Mode (DCM)

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

5.1. Origin of the discontinuous conduction mode, and mode boundary

continuous conduction mode (CCM)

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Reduction of load current

Chapter 5: Discontinuous conduction mode

Further reduce load current

Discontinuous conduction mode

Chapter 5: Discontinuous conduction mode

This expression is of the form

K < K crit(D) for DCM K = 2L and K crit(D) = D' RTs

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

K and Kcrit vs. D

K < Kcrit: DCM

K > Kcrit: CCM

K > Kcrit: CCM K = 2L/RTs

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Critical load resistance Rcrit

Solve Kcrit equation for load resistance R:

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Summary: mode boundary

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Analysis of the conversion ratio M(D,K)

Analysis techniques for the discontinuous conduction mode:

Capacitor charge balance

Small ripple approximation sometimes applies:

Chapter 5: Discontinuous conduction mode

Example: Analysis of DCM buck converter M(D,K)

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

L + vL(t) iC(t) R + v(t)

vL(t) = Vg v(t) iC(t) = iL(t) v(t) / R

Small ripple approximation for v(t) (but not for i(t)!):

vL(t) Vg V iC(t) iL(t) V / R

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

vL(t) = v(t) iC(t) = iL(t) v(t) / R

Small ripple approximation for v(t) but not for i(t):

vL(t) V iC(t) iL(t) V / R

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

vL = 0, iL = 0 iC(t) = iL(t) v(t) / R

L + vL(t) C iC(t) R + v(t)

Small ripple approximation:

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Inductor volt-second balance

vL(t) = D1(Vg V) + D2( V) + D3(0) = 0

note that D2 is unknown

Fundamentals of Power Electronics

Chapter 5: Discontinuous conduction mode

Capacitor charge balance