DC/DC Converter Solving Methodology II: Ripple in State Variables
DC/DC Converter Solving Methodology II: Ripple in State Variables
DC/DC Converter Solving Methodology II: Ripple in State Variables
1, 1-5, 2013
Additional note
Abstract
This document extends the basic methodology of DC7DC converters, in order to calculate ripple current in inductors, and
ripple voltage in capacitors.
Keywords
Power Electronics — Methodology
1 Departmento de tecnologı́a Electrónica, UPV/EHU, E.T.S.I. de Bilbao
*email: [email protected]
1 1
Contents • diL (t) = vL (t)dt dvC (t) = iC (t)dt
L C
Introducción 1 Whenever the voltage in an inductor is constant, the
1 Methodology 1 current changes linearly.
2 Recomended readings 4 Whenever the current in a capacitor is constant, the
current changes linearly.
3 Exercises 5
Z i2 Z t2 Z v2 Z t2
1 1
• diL (t) = vL (t)dt dvC (t) = iC (t)dt
i1 L t1 v1 C t1
Introduction
Z t2 Z t2
The goal of this document is to extend the DC/DC converter 1 1
• ∆iLt1−t2 = vL (t)dt ∆vCt1−t2 = iC (t)dt
solving methodology to calculate ripple in state variables. L t1 C t1
vL (t) iC (t)
1. Methodology
In the previous section a method for calculation of state vari- A+ A+
ables in a DC/DC converter Ewas presented. The method used
the concept of ”small ripple approximation”, and the ripple in
inductor currents and capacitor voltages was not considered. A− A−
In reality inductor current and capacitor voltage present
ripple, and this ripple is relevant in the design or selection of
iL (t) vC (t)
these components and in the quality of the power converter.
In the following lines, a simple method to calculate these
variations in the state variables is described. The method is ∆vC ∆vC
∆iL ∆iL
based on the graphical representation of inductor voltage and
capacitor current in a switching period.
The current rises when positive voltage is applied to an
inductor, and it will drop when negative voltage is applied. Previous figures show how positive areas make the state
Similarly, voltage rises when positive current is applied to a variables rise, and negative areas make them fall. It can also be
capacitor, and it will drop when negative current is applied. apreciated that rectangular areas (constant voltage in inductor
if voltage in an inductor or current in a capacitor is pre- or constant current in capacitor) lead to constant slope rise/fall
sented graphically, it is straightforward to detect the tiem in the state variable.
intervals when the variable are increasing or decreasing, and Within a switching period, maximum variation of the state
their maximum and minimum values, as shown in next equa- variable takes place in the interval [t1 ,t2 ] when the absolute
tions: difference between positve and negative areas is maximum.
vL (t)
Example circuit
An example circuit (Buck converter) is used to calcu- vL (t) = Vdc −VC
late ripple in the state variables. hvL (t)iT = 0 =⇒ A+ = A−
The solution requires solving previously the av- A+ To f f
erage values of state variables, as described in the
preceding section. t
A−
Ton
L vL (t) = −VC
C R
The graphic must clearly indicate positive and negative
areas.
Inductor current waveform is shown in next figure: From the preceding section
vL (t) VR = VC = DVdc
vL (t) = Vdc −VC the following result is obtained
A+ 1
∆iLmax =≈ (1 − D)DVdc (10)
t L fsw
Ton
vL (t) = −VC
1
−∆iLmax ≈ − (1 − D)DVdc (11)
L fsw
iL (t)
If a high enough switching frequency or inductor value
∆iL
is used, the ripple can be reduced to the desired value.
¿What is more interesting, increasing L or fsw ? ¿What
t
drawbacks present both alternatives?
1 + (Vdc −VC ) × Ton 1
∆iLmax = Amax = = (1 − D)DVdc Up to this point, it is time to reflect on the results obtained in
L L L fsw the las two sections. The preceding section gave as a constan
Maximum variation of inductor current is
value for the current in inductor, and this section provides a
Z t2 ripple value for this variable. The first step provides the
1 average value of the inductor current, and the secon step
∆iLmax = ILmax − ILm in = vL (t)dt (5)
L t1 provides the exact waveform and ripple around this average
When capacitor voltage ripple is neglected, calculation of value.
inductor ripple generally results in very simple expressions
3th step: Graphical representation of current in ca-
because at each state of the converter, capacitor voltage is
pacitors and detection of positive and negative ar-
almost constant. From the graphic of inductor voltage,
eas in a switching interval.
maximum difference in areas takes place between 0 and Ton
Unlike with inductor voltage, the reader can represent
(positive area and rising slope) or between Ton and T
graphically current in capacitor in a switching interval, Tsw ,
(negative area and falling slope). Between any other two
from the data obtained in the previous steps of the proposed
intervals within the switching period the area is not maximum
method, ”and the real inductor current waveforms obtained
and thus, variation in current is not maximum.
by calculation of inductor current ripple”. This ripple can be
neglected, but more accurate results are obtained if it is
Average value of vL (t) must be zero, and both areas
considered during the calculations.
are equal
Example circuit: 3th step. In the case of the Buck
The ripple in current is: converter, the current in the capacitor was obtained in the first
part of the methodology, for clarity the results are repeated
• Interval 0 to Ton : below:
Voltage in inductor is:
Z Ton
1 • During Ton
∆iLmax = ILmax − ILm in ≈ (Vdc −VR )dt (6)
L 0
iL (t) Vdc /R
2. Recomended readings
This document is based on the following documents:
Introduction to solid state power electronics. Appendix I,
iC (t) página A-1. Powerex,
t www.pwrx.com/pwrx/app/solid_st.pdf
The graphic must clearly indicate positive and negative areas. “Fundamentals of Power Electronics”. R.W. Erickson, Kluwer
Academic Publisher, in chapters 2 and 3.
Slides of the book are available at the following links
http://ecee.colorado.edu/˜ecen5797/
4th step: capacitor ripple voltage calculation course_material/Ch2handouts.pdf and
The variation of the voltage is obtained by integration of the http://ecee.colorado.edu/˜ecen5797/
capacitor equation: course_material/Ch3handouts.pdf .
Z t+T0
dvC (t) 1
iC (t) = L =⇒ ∆vC (t) = iC (t)dt (14)
dt C T0
iL (t) Vdc /R
iC (t)
t
vC (t)
1 +
∆vCmax = A
C max
t
3. Exercises
C1
L1
L2 C2 R
−
Vdc +
L2
C1
L1
C2 Vout
− −
Vdc + Vdc +
−
Vdc +
L1
L2 C2 Vout
IDC IDC
C1
− − C1
Vdc + Vdc + L1
L1 L2
1
L2 C2 Vout
+ vL1 − +
2
+ vL2 − +
− VC1 C1
Vs + C2 Vo
2 −
1 − C1
L1 L2
D C2 Vout
5V C R
DC/DC Converter solving methodology II: Ripple in state variables — 6/6
L1 D1 L2
CD
12 C2 Vout
L2
D1
L1 C1 D2 C2 Vout