AN1477

Digital Compensator Design for LLC Resonant Converter

Author: Meeravali Shaik and Ramesh Kankanala INTRODUCTION

Microchip Technology Inc. The LLC resonant converter topology, illustrated in
Figure 1, allows ZVS for half-bridge MOSFETs, thereby
ABSTRACT considerably lowering the switching losses and improv-
ing the converter efficiency. The control system design
A half-bridge LLC resonant converter with Zero Voltage of resonant converters is different from the conven-
Switching (ZVS) and Pulse Frequency Modulation tional fixed frequency Pulse-Width Modulation (PWM)
(PFM) is a lucrative topology for DC/DC conversion. A converters. In order to design a suitable digital
Digital Signal Controller (DSC) provides component compensator, the large signal and small signal models
cost reduction, flexible design, and the ability to monitor of the LLC resonant converter are derived using the
and process the system conditions to achieve greater EDF technique.
stability. The dynamics of the LLC resonant converter
are investigated using the small signal modeling tech-
nique based on Extended Describing Functions (EDF)
methodology. Also, a comprehensive description of the
design for the compensator for control of the LLC
converter is presented.

FIGURE 1: LLC RESONANT CONVERTER SCHEMATIC

Q1
C1
D1

Vin D3
ir isp Io
A
Ls +
Q2 Cs im
rc
R V0
ns
C2 Lm Cf
np –
D2

ns
D4

B
R1
Driver PWM Digital
Compensator ADC
Transformers Out put

Q1 R2

Q2 dsPIC33FJXXGSXXX

 2012 Microchip Technology Inc. DS01477A-page 1

AN1477
Conventional methods, such as State-Space Averaging 3. Extended Describing Function (EDF)
(SSA), have been successfully applied to PWM switch- A linear, stationary system responds to a sinusoid
ing converters. In PWM switching converters, the with another sinusoid of the same frequency, but
switch network is replaced by an average circuit model with modified amplitude and phase. The
and only low-frequency (DC) components are consid- describing function method is used to represent a
ered while ignoring switching harmonics. In general, nonlinear function in a linear manner by
the large and small signal modeling of PWM switching considering only the fundamental component of
converters is done by considering the output LC filter. the response of the nonlinear system.
Typically, the natural frequency (fo) of the output LC
filter is much lower than the switching frequency (fs). In this application note, higher order harmonics
are ignored as they are considered to be
In frequency controlled resonant converters, switching negligible. This principle of describing functions is
frequency is close to the natural frequency of the LC extended to model resonant converters and it is
resonant tank. The inductor current and capacitor volt- labelled as EDF.
age of the LC resonant tank, magnetizing current and
primary voltage of the transformer, contain switching Using the EDF method, the discontinuous terms in
frequency harmonics which must be considered to the nonlinear state equations are approximated to
obtain an accurate model. Therefore, modeling is done their fundamental or DC components.
by considering magnetizing inductance (Lm), leakage 4. Harmonic Balance
inductance (Ls) and resonant capacitance (Cs). The Ls,
The quasi-sinusoidal terms and the nonlinear
Lm and Cs constitute the primary resonant components.
discontinuous terms obtained from the harmonic
The small signal modeling approach, based on the approximation and EDF are substituted in the
EDF method, is generally applied to model LLC reso- state equations. The coefficients of DC, sine and
nant converters as this method considers all switching cosine components are then separated to obtain
frequency harmonics for accuracy. Using the EDF, it is the modulation equations (an approximate large
easy to obtain the commonly used transfer functions, signal model).
such as control-to-output transfer function (Gvω) and
5. Obtaining the Steady-State Operating Point
line-to-output transfer function (Gvg).
A large signal model from the harmonic balance is
used to obtain the steady-state operating point by
SMALL SIGNAL MODELING OF LLC setting the derivative terms of harmonic balance
RESONANT CONVERTER equations to zero. This is because the state
variables do not change with time in steady state.
Resonant DC/DC converters are nonlinear systems
and a dynamic model is helpful to determine the linear- 6. Perturbation and Linearization of Harmonic
ized small signal model, and thereby, the system Balance Equations
transfer functions for the Pulse Frequency Modulated The large signal model obtained from the
DC/DC converters. harmonic balance has nonlinear terms arising
The following seven-step process describes how to from the product of two or more time varying
obtain the plant transfer functions for the PFM DC/DC quantities. The linearized model is obtained by
converters. perturbing the large signal model equations about
a chosen operating point, and by eliminating the
1. Time Variant Nonlinear State Equations
higher order (nonlinear) terms.
State equations are obtained by writing the circuit
7. State-Space Model
equations using Kirchhoff's Laws for each state
variable. The state-space model of a continuous time
dynamic system can be obtained from the
2. Harmonic Approximation
perturbed and linearized model of the harmonic
Quasi-sinusoidal current and voltage waveforms balance equations, described in Step 6, to derive
of the LLC resonant tank are resonant current the control-to-output transfer function.
(ir(t)), magnetizing current (im(t)) and voltage
across resonant capacitor (vcr(t)). These
parameters are approximated to their fundamental
components. The current and voltage of the output
filter are approximated to their DC components.

DS01477A-page 2  2012 Microchip Technology Inc.

 AN1477
Derivation of Nonlinear State Equations
A quasi-square wave voltage (vAB ), generated from the
active half-bridge network, is applied to the resonant
tank of the LLC resonant converter, as illustrated in
Figure 2.

FIGURE 2: EQUIVALENT CIRCUIT OF LLC RESONANT CONVERTER

rs Ls ir Cs ip np: ns : ns isp Io
A
+
+ Vcr – im D3 rc
R Vo
Cf Vcf
–
Lm V'cf
+ VAB
Vin –

D4
Ts
B

The state equations are obtained in Continuous Tank EQUATION 3: TRANSFORMER PRIMARY
Current mode by using Kirchhoff’s Circuit Laws (KCL), VOLTAGE
as shown in Equation 1 through Equation 4.
di m
v' c = L m --------
EQUATION 1: RESONANT TANK VOLTAGE
f dt
di r
v AB = L s  ------- + i r r s + v cr + sgn ( i p )v'c f EQUATION 4: TRANSFORMER
dt
SECONDARY CURRENT
Where:
r dv c 1
sgn(ip) = { -1, if v’cf < 0 i sp =  1 + ---c- C f ---------f + --- v c
 R dt R f
+1, if v’cf ≥ 0}

The output voltage (v0) is shown in Equation 5.

In this application, the LLC resonant converter output
voltage is regulated by modulating the switching
frequency (ωs). EQUATION 5: OUTPUT VOLTAGE
r' c
v o = r' c × abs ( i sp ) +  ----- v c
EQUATION 2: RESONANT TANK rc f

CURRENT Where:
r' c = r c R
dv cr
i r = C s ----------
dt

 2012 Microchip Technology Inc. DS01477A-page 3

AN1477
Applying Harmonic Approximation
The Fourier series decomposes periodic functions or
periodic signals into a sum of (possibly infinite) simple
oscillating functions (sines and cosines, or complex
exponentials). Expressing the function (f(x)) as an infi-
nite series of sine and cosine functions is shown in
Equation 6.

EQUATION 6: GENERAL FOURIER EXPANSION

∞
f ( x ) = a0 ±  ( an sin nx + bn cos nx )
n=1
= ( a 0 ± a 1 sin x ± a 2 sin 2x ± a 3 sin 3x ± b 1 cos x ± b 2 cos 2x ± b 3 cos 3x )
Expressing f(x) by considering only the fundamental components and ignoring the DC component, and other
harmonic terms is :
f ( x ) = a 1 sin x ± b 1 cos x

The primary side resonant tank parameters, ir(t), vc(t) The parameters, sine component of resonant
and im(t), provided in Equation 7, are approximated to current (is), cosine component of resonant current (ic),
their fundamental harmonics, and the output filter volt- sine component of resonant capacitor voltage (vs),
age (vcf) is approximated to the DC component. The cosine component of resonant capacitor voltage (vc),
derivatives of ir(t), vcr(t) and im(t) are shown in sine component of magnetizing current (ims) and cosine
Equation 7. component of magnetizing current (imc) are slow time
varying components. Therefore, the dynamic behavior
EQUATION 7: FUNDAMENTAL of these parameters can be analyzed.
APPROXIMATION OF Figure 3 and Figure 4 illustrate the simulation wave-
PRIMARY TANK forms of the LLC resonant converter operating below
PARAMETERS the resonant frequency and continuous tank current
mode.
i r ( t ) = i s ( t ) sin ω s t – i c ( t ) cos ω s t

v c r ( t ) = v s ( t ) sin ω s t – v c ( t ) cos ω s t

i m ( t ) = i ms ( t ) sin ω s t – i mc ( t ) cos ω s t

di r di di
------- =  -------s + ω s i c sin ω s t –  -------c – ω s i s cos ω st
dt  dt   dt 

dvc r dv dv
- =  --------s + ω s v c sin ω s t –  --------c – ω s v s cos ω s t
----------
dt  dt   dt 

di m di m s di mc
--------- =  ----------
- + ωs i mc sin ωs t –  -----------
- – ω s i m s cos ω s t
dt dt dt

Where:
ωs = switching frequency in radians/second

DS01477A-page 4  2012 Microchip Technology Inc.

 AN1477
FIGURE 3: SIMULATION WAVEFORMS OF LLC RESONANT CONVERTER
Input Voltage

Time (ms)

Resonant Inductor Current

Time (ms)

Resonant Capacitor Voltage

Time (ms)

 2012 Microchip Technology Inc. DS01477A-page 5

AN1477
FIGURE 4: SIMULATION WAVEFORMS OF LLC RESONANT CONVERTER

Magnetizing Inductor Current

Time (ms)

Output Filter Capacitor Voltage

Time (ms)

DS01477A-page 6  2012 Microchip Technology Inc.

 AN1477
Applying Extended Describing The fundamental output voltage of a half-bridge
Function (EDF) inverter is shown in Equation 9.

Extended Describing Function is a powerful mathemat- EQUATION 9: OUTPUT VOLTAGE OF

ical approach for understanding, analyzing, improving HALF-BRIDGE INVERTER
and designing the behavior of nonlinear systems.
Every system is nonlinear, except in limited operating (π – θ)
regions. 2
f 1 ( d, v in ) = ------
2π  v in × sin ( ωt ) dωt
The nonlinear terms provided in Equation 1 through θ
Equation 5, sgn(ip) * vcf’ and abs(isp) can be approxi- 2v in (π – θ)
mated to their fundamental harmonic terms and DC f 1 ( d, v in ) = – ----------- cos ( ωt )
2π θ
terms.
2v in
The functions, f1(d, vin ), f2(iss, isp,v’cf ), f3(isc , isp, v’cf ) and f 1 ( d, v in ) = ----------- [ cos θ – cos ( π – θ ) ]
f4(iss, isc ), are called EDFs. Where, iss, isc are the sine and 2π
cosine components of the transformer secondary cur- 2v in 2v in
π dπ
rent, and isp is the resultant current flowing in secondary. f 1 ( d, v in ) = ----------- cos θ = ----------- × cos  --- – ------
π π 2 2 
f1, f2, f3 and f4 are functions of the harmonic coefficients
of state variables at chosen operating conditions. The 2v in π
f 1 ( d, v in ) = ----------- sin  --- d = ves
EDF terms can be calculated by using the Fourier π 2 
expansion of nonlinear terms. The EDF approximation
to nonlinear states is shown in Equation 8. Where:
θ = π--- – dπ
------
EQUATION 8: EDF APPROXIMATION 2 2
v AB ( t ) = f 1 ( d, v in ) sin ωs t ves = Sine component of the output voltage
of half-bridge inverter

sgn ( i sp ) v' c = f 2  i ss, i sp, v'c  sin ω s t – f 3  i sc, i sp, v'c  cos ω s t
f  f  f
The switching waveform has an odd symmetry. There-
i sp = f 4 ( i ss, i sc ) fore, there is no cosine component (vec = 0, where vec
is the cosine component of the output voltage of the
Figure 5 illustrates a typical switching waveform of a half-bridge inverter) in the switching waveform and the
half-bridge inverter which is the input to the LLC sine component (ves) forms the fundamental
resonant tank (θ = Dead Time, d = Duty Cycle). component of vAB .

FIGURE 5: OUTPUT SWITCHING

WAVEFORM OF
HALF-BRIDGE INVERTER

Vin

θ θ
dπ

0 π 2π
ωt

 2012 Microchip Technology Inc. DS01477A-page 7

AN1477
The EDF approximation to the nonlinear transformer Substituting Equation 9 and Equation 10 into
primary voltage is shown in Equation 10. Equation 1 through Equation 5, and separating the DC,
sine and cosine terms, Equation 11 through
EQUATION 10: EDF APPROXIMATION TO Equation 13 are obtained.
TRANSFORMER PRIMARY
VOLTAGE EQUATION 11: SINE AND COSINE
4 ip s
COMPONENTS OF TANK
4 i ss
f 2  i ss, i sp, v'c  = --- ------ v' c = --- -------- v' c VOLTAGE
 f π i sp f π i pp f
di
4n i ps v es = L s  -------s + ω s i c + r s i s + v s + v ps
 dt 
= ------ -------- v c = v p s
π i pp f
di 4n i ps
= L s  -------s + ω s i c + r s i s + v s + ------ -------- v c
 dt  π i pp f
4 i sc 4 i pc
f 3  i sc, i sp, v'c  = --- ------ v' c = --- -------- v' c di
 f π i sp f π i pp f
v ec = L s  -------c – ω s i s + r s i c + v c + v pc
 dt 
4n i pc
= ------ -------- v c = v pc di c 4n i p c
π i pp f = L s  ------- – ω s i s + r s i c + v c + ------ -------- v c
 dt  π i pp f
2 2
i pp = i p s + i pc

Where: EQUATION 12: SINE AND COSINE

vps, vpc = sine, cosine components of the COMPONENTS OF TANK
transformer primary voltage CURRENT
ips, ipc = sine, cosine components of the dv
i s = C s  --------s + ω s v c
transformer primary current  dt 
ipp = resultant transformer primary current dv c
i c = C s  -------- – ω s v s
iss, isc = sine, cosine components of the dt
transformer secondary current
isp = resultant current flowing in secondary EQUATION 13: SINE AND COSINE
n = np/ns = transformer turns ratio COMPONENT OF
TRANSFORMER PRIMARY
VOLTAGE
Harmonic Balance di m s 4n i p s
L m  ----------
- + ω s i mc = ------ -------- v c = v p s
 dt  π i pp f
Harmonic balance is a frequency domain method used
to calculate the steady-state response of nonlinear di mc i pc
differential equations. The term, “harmonic balance”, is L m  ------------ – ω i  = 4n -v = v
------ -------
 dt s ms π i pp c f pc
descriptive of the method, which uses the Kirchhoff's
Current Laws (KCL) written in the frequency domain
and a chosen number of harmonics. Effectively, the Only the DC term is considered for the output capacitor
method assumes that the solution can be represented voltage, as shown in Equation 14.
by a linear combination of sinusoids, and then balances
current and voltage sinusoids to satisfy the Kirchhoff's EQUATION 14: OUTPUT FILTER
Laws. The harmonic balance method is commonly CAPACITOR VOLTAGE
used to simulate circuits which include nonlinear
r dv c
elements.  1 + ----c- C ----------f + --1- v = --2- i
 R  f dt R cf π sp

The output voltage equation is shown in Equation 15.

EQUATION 15: OUTPUT VOLTAGE

2  r' c
v 0 = --- r' c i s p +  ------ v c
π  rc  f

DS01477A-page 8  2012 Microchip Technology Inc.

 AN1477
Equation 11 through Equation 15 are the nonlinear Equation 15, {vg, ωs, d}, is slow varying quantities with
large signal model of the LLC resonant converter respect to the switching frequency. Therefore, the
power stage and are illustrated in Figure 6. It is impor- modulation equations can be easily perturbed and
tant to note that the input of Equation 12 through linearized at chosen operating points.

FIGURE 6: LARGE SIGNAL MODEL OF LLC RESONANT CONVERTER

is rs Ls + Vs - i ps
– +
Cs i ms
Ωs Ls ic Lm Io
+ + Vps
Ves Ωs Cs Vc – +
–
– rc
Ωs Lmi mc +
R
2/p isp Vo
Ω s Lmi ms – Cf Vcf
Vec – + – –
Ω s Cs V s
+ + Vpc
Ω s Ls is Lm
ic rs Ls imc ipc
Cs
+ –
+ Vc –

Deriving Steady-State Operating Point The output filter capacitor voltage can be calculated by
substituting Equation 16 into Equation 14, as shown in
Under steady-state conditions, the state variables of Equation 17.
the modulation equations, Equation 12 through
Equation 14, do not change with time. For a chosen
EQUATION 17: FILTER CAPACITOR
operating point, the time derivatives in Equation 12
VOLTAGE
through Equation 14 are set to zero and the steady-
vc
state values are obtained (shown in upper case letters).
------f = ------ i pp = --- i s p
2n 2
The transformer currents on the primary and secondary R π π
sides are shown in Equation 16. 2n
 v c = ------ i pp R
f π
EQUATION 16: TRANSFORMER CURRENTS π
 v c = ------ i pp R e
Primary Current: f 4n
2 2 π
i pp = ip s + ip c V c' f = nV c = --- I pp R e
f 4
8 2
Secondary Current: R e = -----2 n R = equivalent load resistance referred
π
2 2 2 2 to primary side
i sp = i ss + i sc = n i ps + i pc = ni pp
Where:
Where: V 'c f = reflected voltage of secondary on the
n = np/ns = transformer turns ratio primary

The steady-state analysis for the tank current, resonant

capacitor voltage and magnetizing current are provided
in Equation 18 through Equation 22.

 2012 Microchip Technology Inc. DS01477A-page 9

AN1477
Substituting the value of Equation 17 into the sine Substituting the value of Equation 17 into the sine
component of tank voltage, the result obtained is component of magnetizing current, the result is shown
shown in Equation 18. in Equation 21.

EQUATION 18: SINE COMPONENT OF EQUATION 21: SINE COMPONENT OF

TANK VOLTAGE MAGNETIZING CURRENT
di 4 ip s di ms 4n i p s
v e s = L s  -------s + ω s i c + r s i s + v s + --- -------- v' c L m  ---------- + ωs i mc = ------ × -------- × v c
 dt  π i pp f  dt  π i pp f
I
4 ps π 2
 L s Ω s I c + r s I s + V s + --- -------- --- I pp Re = Ve s = --- V in  L m Ω s I mc – R e I p s = 0
π I pp 4 π
 R e I s – L m Ω s I mc – R e I ms= 0
2
 r s I s + L s Ω s I c + V s + R e I p s = --- V in
π
Substituting the value of Equation 17 into the cosine
2
 ( r s + R e )I s + L s Ω s I c + V s – R e I ms = --- V in component of magnetizing current, the result is shown
π
Where: in Equation 22.
Ip s = Is – Im s
EQUATION 22: COSINE COMPONENT OF
MAGNETIZING CURRENT
Substituting the value of Equation 17 into the cosine
component of the tank voltage, the result obtained is di mc 4n i pc
shown in Equation 19. L m  -----------
- – ω s i ms = ------ -------- v c
dt π i pp f

EQUATION 19: COSINE COMPONENT OF  ( – L m Ω s I ms ) – R e I pc = 0

TANK VOLTAGE  L m Ω s I ms + R e I c – R e I m c = 0
di c 4 i pc
v ec = L s  ------- – ω s i s + r s i c + v c + --- -------- v' c
 dt  π i pp f Equation 19 through Equation 22 are arranged, as
4 I pc π shown in Equation 23.
 – L s Ω s I s + r s I c + V c + --- -------- --- I pp R e = V ec = 0
π I pp 4
EQUATION 23: ARRANGEMENT OF
 – L s Ω s I s + r s I c + V c + I pc R e = 0 STEADY-STATE
EQUATIONS
 – L s Ω s I s + ( r s + R e )I c + V c – I mc R e = 0
Where: 2
( r s + R e )I s + L s Ω s I c + V s – R e I m s = --- Vin = Ves
I pc = I c – I mc π
– L s Ω s I s + ( r s + R e )I c + V c – I mc R e = 0 = V ec
The steady-state values of sine and cosine compo- Is – Cs Ωs V c = 0
nents of the tank current can be obtained by equating
dvs/dt and dvc /dt to zero. The result is shown in Ic + Cs Ωs Vs = 0
Equation 20. R e I s – L m Ω s I mc – R e I m s = 0

EQUATION 20: SINE AND COSINE L m Ω s I m s + R e I c – R e I mc = 0

COMPONENTS OF TANK
CURRENT
dv
C s  -------s- + ω s v c = i s
 dt 
 Is – Cs Ωs Vc = 0
dv
C s  -------c- – ω s v s = i c
dt
Ic + Cs Ωs Vs = 0

DS01477A-page 10  2012 Microchip Technology Inc.

 AN1477
To obtain the tank current, capacitor voltage and mag- Perturbation and Linearization of
netizing current from the steady-state equations, Harmonic Balance Equations
Equation 23 is formulated in the matrix form, as shown
in Equation 24. The nonlinear system equations, Equation 12 through
Equation 14, are in the form of: x′ = f(x(t), u(t));
EQUATION 24: STEADY-STATE x(t) = state of the nonlinear system and u(t) = input to
OPERATING POINT the system.
X × Y = U0 The function (x′) can be linearized about an operating
point and is expressed in the form of: x′ = Ax + Bu,
–1 where A and B are the Jacobian matrices of the system
Y = X × U0
Where: with respect to x(t) and u(t), as shown in Equation 25.
rs + Re Ls Ωs 1 0 –Re 0
EQUATION 25: JACOBIAN MATRICES
–Ls Ω s rs + Re 0 1 0 –Re
δf (x ( t ),u ( t ))
1 0 0 – Cs Ω s 0 0 A ij = --------------------------------i
X = ∂x j ( t )
0 1 Cs Ω s 0 0 0 x 0, u 0
δf (x ( t ),u ( t ))
Re 0 0 0 – R e – Lm Ω s B ij = --------------------------------i
∂u j ( t )
0 Re 0 0 Lm Ω s –Re x 0, u 0
Where:
Is x0 and u0 represent the steady-state operating
V es
Ic points.
0
Vs
U0 = 0
Y = In the perturbation and linearization step, it is assumed
0 Vc that the averaged state variables and the input vari-
0 I ms ables consist of the constant DC component and a
0 small signal AC variation about the DC component.
I mc Perturbed signals are shown in Equation 26.

EQUATION 26: PERTURBED SIGNALS

v in = V in + v̂ in , d = D + dˆ , ω s = Ω s + ω̂ s , v = V + v̂ ps , v
ps ps pc = V pc + v̂ pc , i p s = I p s + î p s ,

i pc = I pc + î pc , i
pp = I pp + î pp , v c = V c + v̂ c , i m s = I m s + î m s , i mc = I mc + î mc , i s = I s + iˆs ,
f f f

i c = I c + î c , v s = V s + v̂ s , v c = V c + v̂ c , v e s = V e s + v̂ e s , v 0 = V 0 + v̂ 0

 2012 Microchip Technology Inc. DS01477A-page 11

AN1477
The sine component of the transformer primary
voltage (vps) is linearized around the steady-state
operating point, as shown in Equation 27.

EQUATION 27: LINEARIZATION OF SINE COMPONENT OF TRANSFORMER PRIMARY VOLTAGE

4n i p s 4n ip s
v p s = ------ -------- v c = ------ -------------------------------- v c
π i pp f π 2 2 f
i ps + i pc
 I 2 + I 2 – ------------------------------- 1
-
 ps pc 2 2 
4nV c  I + I  4nV c I p s I pc
f ps pc f I ps
v̂ p s = -------------- ×  ----------------------------------------------------------------------- î ps – -------------- × ------------------------------------3- î p c + ------ × ------------------------------ × v̂ c
4n
π  2 2  π --- π f
 I ps + I pc  2 I ps2 + I pc2
   I 2 + I 2
   ps pc 

4nV c 2 4nV c
f I pc f I ps I pc- 4n I ps
v̂ ps = -------------- ---------3 î ps – -------------- -------------- î p c + ------ ------- v̂ c
π I π I pp
3 π I pp f
pp

v̂ ps = H ip î ps + H ic î p c + H vcf v̂ c
f

v̂ ps = H ip î s + H ic î c – H ip î'ims
ms – H ic î mc + H vcf v̂ c f
Where:
4nV c 2
f I pc
H ip = -------------- ---------3
π I
pp
4nV c
f I ps I pc-
H ic = – -------------- --------------
π I pp
3

4n I ps
H vcf = ------ -------
π I pp

DS01477A-page 12  2012 Microchip Technology Inc.

 AN1477
The cosine component of the transformer primary
voltage (vpc) is linearized around the steady-state
operating point, as shown in Equation 28.

EQUATION 28: LINEARIZATION OF COSINE COMPONENT OF TRANSFORMER PRIMARY VOLTAGE

4n i pc 4n i pc
v pc = ------ × ------ × v c = ------ × ------------------------------ × v c
π i pp f π 2 2 f
i ps + i pc

 I 2 + I 2 – ------------------------------ 1 
   ps pc 
  4nV c  I I  4nV c  I
2
+ I
2  I pc 
f f
=   – -------------- ---------------------------------3- î ps + --------------  -------------------------------------------------------------------
ps pc ps pc  4n  ------------------------------ v̂
v̂ pc î + -----
-
 π  --- π  2  pc π   cf
 2 2 2   I 2
+ I
2    I ps2 + I pc2 
 ( I ps + I pc )    ps pc  
 
4nV c 4nV c 2
f I ps I pc- f I----------- I pc
v̂ pc = – -------------- --------------- î + -------------
- pc
- î pc + 4n
------ -------- v̂
π I pp
3 ps π I pp
3 π I pp c f

v̂ pc = G ip î s + G ic î c – G ip î ms – G ic î mc + G vcf v̂ cf

Where:
4nV c
f I ps I pc-
G ip = – -------------- ---------------
π I pp
3

4nV c I 2
f psc -
G ic = -------------- -----------
π I pp
3

4n I pc
G vcf = ------ -------
π I pp

The linearization of the input voltage (ves) is shown in Removing the steady-state terms and other higher
Equation 29. order perturbed terms in Equation 29 to get the
linearized input voltage is shown in Equation 30.
EQUATION 29: LINEARIZATION OF
HALF-BRIDGE INVERTER EQUATION 30: LINEARIZATION OF
OUTPUT VOLTAGE HALF-BRIDGE INVERTER
2v in OUTPUT VOLTAGE
π
v es = ---------- sin  --- d
π 2  π π
v̂ es = --- sin  --- D v̂ in + V in cos  --- D dˆ
2
π 2 2
π
v̂ es = --- × ( V in + v̂ in ) sin  --- ( D + dˆ )
2
π 2 v̂ es = K 1 v̂ in + K 2 dˆ
π
Expanding sin  --- ( D + dˆ ) Where:
2
π
K 1 = --- sin  --- D
2
π π π π
= sin  --- D cos  --- dˆ  + cos  --- D sin  --- dˆ  π 2 
2  2  2  2 
π
K 2 = V in cos  --- D
2 
π π π
= sin  --- D + --- cos  --- D dˆ
2 2 2

π π π
v̂ es = --- ( V in + v̂ in )  sin  --- D + --- cos  --- D dˆ 
2
π 2 2 2

 2012 Microchip Technology Inc. DS01477A-page 13

AN1477
The linearization and perturbation of the tank current, In resonant converters, the poles and zeroes are
capacitor voltage, transformer primary voltage, output the functions of normalized switching frequency
voltage and output filter capacitor voltage, after remov- (ω sn = ωs/ω0), where ω s is the switching frequency
ing the second order and DC terms, are provided in and ω0 is the resonant frequency.
Equation 31 through Equation 42. The linearization and perturbation of the sine component
of the tank voltage is provided in Equation 31.

EQUATION 31: LINEARIZATION OF SINE COMPONENT OF TANK VOLTAGE

d ( I s + î s )  ω̂ s 
L s ---------------------- + r s ( I s + î s ) + L s ( I c + î c )  Ω s + ω 0 ------- + ( V s + v̂ s ) + ( V ps + v̂ ps ) = ( V es + v̂ es )
dt  ω 0
dî s ˆ
L s ------- + r s î s + Ω s L s î c + L s ω 0 I c ω sn + v̂ s + v̂ ps = v̂ es
dt

Substitute the values of Equation 27 and Equation 30

into the sine component of the tank voltage, as shown
in Equation 32.

EQUATION 32: LINEARIZATION OF SINE COMPONENT OF TANK VOLTAGE

dî s ˆ
L s ------- + r s î s + Ω s L s î c + L s ω 0 I c ω sn + v̂ s + H ip î s + H ic î c – H ip î ms – H ic î mc + H vcf V̂c f = K 1 v̂ in + K 2 dˆ
dt
dî s ˆ
L s ------- = – ( H ip + r s )î s – ( Ω s L s + H ic )î c – v̂ s + H ip î ms + H ic î mc – H vcf V̂ c + K 1 v̂ in + K 2 dˆ – L s ω 0 I c ω sn
dt f
dî  H ip + r s  Ω s L s + H ic H ip H ic H vcf K1 K2 Ls ω0 Ic ˆ
------s- = –  -------------------- î s –  ----------------------------- î c – ----- v̂ s + --------- î ms + --------- î m c – ------------ V̂ c + ------- v̂ in + ------- dˆ – ------------------ ω sn
1
dt  L s   L s  L s L s L s L s f L s L s Ls

The linearization and perturbation of cosine component

of tank voltage is provided in Equation 33.

EQUATION 33: LINEARIZATION OF COSINE COMPONENT OF TANK VOLTAGE

 ˆ
d ( I c + î c ) ωs
L s ---------------------- + r s ( I c + î c ) – L s ( I s + î s )  Ω s + ω 0 ------ + ( V c + v̂ c ) + ( V pc + v̂ pc ) = 0
dt  ω 0
 dî c 
+ Ω s L s î s –
 L s ------- + r s î c – + L s ω 0 I s ω̂ sn + v̂ c + v̂ pc = 0
 dt 

Substituting the values of Equation 28 into the cosine

component of the tank voltage, the result obtained is
shown in Equation 34.

EQUATION 34: LINEARIZATION OF COSINE COMPONENT OF TANK VOLTAGE

 dî c 
+ Ω s L s î s +
 L s ------- + rs î c – – L s ω0 I s ω̂ sn + v̂ c + G ip î s + G ic î c – G ip î ms – Gic î mc + G vcf v̂ c = 0
 dt  f
dî c
L s ------- = ( Ω s L s – G ip )î s – ( G ic + r s )î c – v̂ c + G ip î ms + G ic î mc – G vcf v̂ c + L s ω 0 I s ω̂sn
dt f

dî c ( Ω s L s – G ip ) ( G ic + r s ) 1 G ip G ic G vc f Ls ω0 Is
------- = ------------------------------- î s – ----------------------- iîcc – ----- v̂ c + -------- î ms + -------- î mc – ----------- v̂ c + ----------------- ω̂ sn
dt Ls Ls Ls Ls Ls Ls f Ls

DS01477A-page 14  2012 Microchip Technology Inc.

 AN1477
The linearization and perturbation of the sine component The linearization and perturbation of the cosine
of the tank current is provided in Equation 35. component of the tank current is provided in
Equation 36.
EQUATION 35: LINEARIZATION OF SINE
COMPONENT OF TANK EQUATION 36: LINEARIZATION OF COSINE
CURRENT COMPONENT OF TANK
CURRENT
d ( V s + v̂ s )  ω̂ s 
C s ------------------------- + C s ( V c + v̂ c )  Ω s + ω 0 ------- = ( I s + î s ) d ( V c + v̂ c )  ω̂ s 
dt  ω 0 C s -------------------------- – C s ( Vs + v̂ s )  Ω s + ω 0 ------- = ( I c + î c )
dt  ω 0
dv̂ s
C s -------- + C s Ω s v̂ c + C s ω 0 V c ω̂ s n = î s dv̂ c
dt  C -------- – C Ω v̂  – C ω V ω̂ = î
 s dt s s s s 0 c sn c
dv̂ s 1 Cs Ωs Cs ω0 Vc
-------- = ------ î s – ------------- v̂ c – -------------------- ω̂ s n dv̂ c 1 Cs Ωs Cs ω0 Vs
dt Cs Cs Cs -------- = ------ îics + ------------- v̂ s + -------------------- ω̂ sn
dt Cs Cs Cs

The linearization and perturbation of the sine component

of the magnetizing current is provided in Equation 37.

EQUATION 37: LINEARIZATION OF SINE COMPONENT OF MAGNETIZING CURRENT

d ( I ms + î ms )  ω̂ s 
L m -------------------------------- + L m ( I mc + î mc )  Ω s + ω 0 ------- = ( V ps + v̂ ps )
dt  ω 0
dî ms ˆ
L m ------------ + L m Ω s î mc + L m I mc ω 0 ω sn = v̂ ps
dt

Substituting the value of Equation 27 into the sine com-

ponent of the transformer primary voltage, the results
are shown in Equation 38.

EQUATION 38: LINEARIZATION OF SINE COMPONENT OF MAGNETIZING CURRENT

dî ms ˆ
L m ------------ + L m Ω s î mc + L m I mc ω 0 ω sn = H ip î s + H ic î c – H ip î ms – H ic î mc + H vcf v̂ cf
dt
dî ms ˆ
L m ------------ = H ip î s + H ic î c – H ip î ms – ( H ic + L m Ω s )î mc + H vcf v̂ cf – L m I mc ω 0 ω sn
dt
dî ms H ip H ic H ip ( H ic + L m Ω s ) H vcf L m I mc ω 0 ˆ
- = --------- î s + --------- î c – --------- î ms – ----------------------------------- î mc + ------------ v̂ cf – ------------------------- ω sn
-----------
dt L m L m L m L m L m Lm

 2012 Microchip Technology Inc. DS01477A-page 15

AN1477
The linearization and perturbation of the cosine
component of the tank voltage is provided in
Equation 39.

EQUATION 39: LINEARIZATION OF COSINE COMPONENT OF MAGNETIZING CURRENT

d ( I mc + î mc )  ω̂ s 
L m --------------------------------- – L m ( I ms + î ms )  Ω s + ω 0 ------- = ( V pc + v̂ pc )
dt  ω 0
dî mc ˆ
L m ------------ – L m Ω s î ms – L m I ms ω 0 ω sn = v̂ pc
dt

Substituting Equation 28 into the cosine component of

the magnetizing current, the result is shown in
Equation 40.

EQUATION 40: LINEARIZATION OF COSINE COMPONENT OF MAGNETIZING CURRENT

dî mc ˆ
L m ----------- – L m Ω s î ms – L m I ms ω 0 ω sn = G ip î s + G ic î c – G ip î ms – G ic î mc + G vcf v̂ cf
dt
dî mc G ip G ic ( G ip – L m Ω s ) G ic G vcf L m I ms ω 0 ˆ
ic – ----------------------------------- î ms – --------- î ms c + ---------- v̂ cf + ------------------------- ω sn
- = --------- î s + --------- îc
-----------
dt Lm Lm Lm Lm Lm Lm

The linearization and perturbation of the output filter

capacitor voltage is provided in Equation 41.

EQUATION 41: LINEARIZATION OF OUTPUT CAPACITOR VOLTAGE

d  V c + v̂ c 
r  f f 1 
 1 + ----c- C ----------------------------- - + --- V c + v̂  = --- ( I + î sp )
2
 R  f dt R f c f π sp
r dv̂ c
  1 + ----c- C f × ----------f + --1- v̂ c = --- î sp
2
 R dt R f π
From Equation 16:
2n 2 2
i sp = ------ i ps + i pc
π
2n I ps 2n I pc
î sp = ------ --------------------------------- î ps + ------ --------------------------------- î pc
π 2 2 π 2 2
I ps + I pc I ps + I pc

 K is î ps + K ic î pc

î sp = K is î s + K ic î c – K is î ms – K ic î mc
dv̂ c
rc f
 1 + ----- C f × ---------- = K is î s + K ic î c – K is î ms – K ic î mc –  --- v̂ c
  1
 R dt  R f
Where:
i ps = i s – i m s and i pc = i c – i mc

2n I ps I pc
2n
K is = ------ --------------------------------- and K ic = ------ ---------------------------------
π π
I ps 2 + I pc 2 2
I ps + I pc
2

rc dv̂ c
f
 ------ C f + ---------- = K is î s + K ic î c – K is î ms – K ic î mc –  --- v̂ c
1
r' c dt R f
dv̂ c K is r' c K is r' K is r' c K ic r' c r' c
ic c
----------f = --------------- î s + --------------- î c – --------------- î ms – --------------- î mc – --------------- v̂ c
dt Cf rc Cf rc Cf rc Cf rc RCf rc f

DS01477A-page 16  2012 Microchip Technology Inc.

 AN1477
The linearization and perturbation of the output voltage
is provided in Equation 42.

EQUATION 42: LINEARIZATION OF OUTPUT VOLTAGE

 r' c
V 0 + v̂ 0 = --- r' c ( I sp + î sp ) +  ------  V c + v̂c 
2
π  rc   f f
2  r' c
 v̂ 0 = --- r' c î sp +  ------ v̂ c
π  rc  f
 r' c
 v̂ 0 = r' c ( K is î s + K ic î c – K is î ms – K ic î mc ) +  ------ v̂ c
 rc  f
 r' c
v̂ 0 = ( K is r' c î s + K ic r' c î c – K is r' c î ms – K ic r'ˆc i ) +  ------ v̂ c
mc  r  f
c

Equation 31 through Equation 42 are arranged, as

shown in Equation 43.

EQUATION 43: LINEARIZED SMALL SIGNAL MODEL OF LLC RESONANT CONVERTER

dî  H ip + r s  Ω s L s + H ic H ip H ic H vcf K1 K2 Ls ω0 Ic ˆ
------s- = –  -------------------- î s –  ----------------------------- î c – ----- v̂ s + --------- î ms + --------- î m c – ------------ v̂ cf + ------- v̂ in + ------- dˆ – ------------------ ω sn
1
dt  Ls   Ls  Ls Ls Ls Ls Ls Ls Ls

dî ( Ω s L s – G ip ) ( G ic + r s ) G ip G ic G vcf Ls ω0 Is
-------c = --------------------------------- î s – ------------------------- î c – ----- v̂ c + --------- î ms + --------- î mc – ------------ v̂ cf + ------------------ ω̂ sn
1
dt Ls Ls Ls Ls Ls Ls Ls

dv̂ s Cs Ω C s ω Vc
1 s 0
-------- = ------ î s – -------------- v̂ c – -------------------- ω̂ sn
dt Cs Cs Cs

dv̂ c Cs Ω C s ω Vs
1 s 0
-------- = ------ î c + -------------- v̂ s + -------------------- ω̂ sn
dt Cs Cs Cs

dî ms H ip H ic H ip H ic + L m Ω s H vcf L m I mc ω
0
- = --------- î s + --------- î c – --------- î ms – ------------------------------- î mc + ------------ v̂ cf – ------------------------- ω̂ sn
-----------
dt Lm Lm Lm Lm Lm Lm

dî mc G ip G ic ( G ip – L m Ω s ) G ic G vcf L m I ms ω
0
- = --------- î s + --------- î c – ----------------------------------- î ms – --------- î mc + ------------ v̂ cf + ------------------------- ω̂ sn
-----------
dt Lm Lm Lm Lm Lm Lm

dV̂ C K is r'c K ic r'c K is r' K is r' r' c

-------------f = --------------- î s + --------------- î c – -------------- c- î – --------------
c- î – -------------- - v̂
dt Cf c r Cf c r Cf c r ms Cf c r mc RC f r c cf

The output equation is:

 r'c 
v̂ 0 = K is r'c î s + K ic r'c î c – K is r'c î ms – K ic r'c î mc +  ------ v̂ cf
 rc 

 2012 Microchip Technology Inc. DS01477A-page 17

AN1477
Formation of State-Space Model The state-space representation (known as time domain
approach) provides a convenient and compact way to
State-space representation is a mathematical model of model and analyze systems with multiple inputs and
a physical system as a set of input, output and state outputs.
variables, related by first order differential equations.
Equation 44 provides the state-space representation of
Additionally, if the dynamic system is linear and time the LLC resonant converter.
invariant, the differential and algebraic equations may
be written in matrix form.

EQUATION 44: STATE-SPACE MODEL OF LLC RESONANT CONVERTER

dx̂- = Ax̂ + Bû

-----
dt
ŷ = Cx̂ + Dû
Where:
T
x̂ =  î s î c v̂ s v̂ c î ms î mc v̂ c  States of the system
 f 
ˆ
û = ( f sn or ω̂ sn ) Control inputs and all other disturbance inputs are ignored

ŷ = ( v̂ 0 ) Output

H ip + r s ( Ω s L s + H ic ) H ip H ic H vcf
1 --------- ---------
– -------------------- – --------------------------------- – ----- 0 – ------------
Ls Ls Ls Ls Ls Ls
( Ω s L s – G ip ) G ic + r s G ip G ic G vcf
--------------------------------- 1 --------- ---------
– ------------------- - 0 – ----- – ------------
Ls Ls Ls Ls Ls Ls

1 Cs Ωs
------ 0 0 – -------------
- 0 0 0
Cs Cs

1 Cs Ωs
A = 0 ------ -------------- 0 0 0 0
Cs Cs
H ip H ic H ip H ic + L m Ω s H vcf
--------- --------- 0 0 – -------- - - ------------
– ------------------------------
Lm Lm Lm Lm Lm
G ip G ic G ip – L m Ω s G ic G vcf
--------- --------- 0 0 – ------------------------------
- – --------- ------------
Lm Lm Lm Lm Lm
K is r' c K ic r' c K is r' c K ic r' c r' c
--------------- --------------
- 0 0 – -------------- - – -------------- - – --------------
-
Cf rc Cf rc Cf rc Cf rc RC f rc

T
 L s ω 0 I c  L s ω 0 I s  C s ω 0 V c  C s ω 0 V s  L m ω 0 I mc  L m ω 0 I ms 
B =   – ----------------
- ---------------- – ------------------- ------------------- – ---------------------- ---------------------- 0
 Ls   Ls   Cs   Cs   Lm   Lm  
 r'
 c 
C =  ( K is r' c ) ( K i c r' c ) ( 0 ) ( 0 ) ( – K is r' c ) ( – K ic r' c )  ------ 
  rc  
D = 0
For the linearized system, the required control-to-output voltage transfer function is:

v̂ 0 –1
- = C ( SI – A ) B + D = G p ( s )
---------
ω̂ sn

DS01477A-page 18  2012 Microchip Technology Inc.

 AN1477
HARDWARE DESIGN Equation 44 can be solved using MATLAB® to obtain
the control-to-output (plant) transfer function,
SPECIFICATIONS sys = ss(A, B, C, D). The ss command arranges
Series resonant inductor (Ls) = 62 µH the A, B, C and D matrices in a state-space model. The
Gp(s) = tf(sys) command gives the transfer function
Series resonant capacitance (Cs) = 9.4 nF
of the system, where sys indicates the system. The
Magnetizing inductor (Lm) = 268 µH plant transfer function (Gp(s)), along with the design
Input voltage (Vin) = 400V (DC) values, are shown in Equation 45.
Output filter capacitance (Cf ) = 2000 µF R=1/Re
Output power = 200W
Switching frequency (fs) = 200 kHz
DCR of resonant inductor (rs) = 15 mΩ
ESR of output capacitor (rc) = 15 mΩ

EQUATION 45: PLANT TRANSFER FUNCTION

4 5
224213315399 × ( s + 3.314 × 10 ) × ( s – 8.262 × 10 )
G p ( s ) = --------------------------------------------------------------------------------------------------------------------------------------------------------------
2 8 2 5 12
( s + 973.6s + 8.949 × 10 ) × ( s + 2.76 × 10 s + 1.227 × 10 )

 s   s 
5.5895  1 + ---------------------------- - ×  ---------------------------- - – 1
 3.314 × 10   8.262 × 10
4 5 
G p ( s ) = ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 2   2 5 
s 973.6s s 2.75 × 10 s
 ----------------------------- + ----------------------------- + 1 ×  -------------------------------- + -------------------------------- + 1
 8.949 × 10 8
8.949 × 10
8   1.227 × 10 12
1.227 × 10
12 

The general form of Gp(s) is shown in Equation 46. The [p, z] = pzmap (Gp (s)) command gives the poles
and zeros of the plant transfer function. Figure 7 illus-
EQUATION 46: GENERALIZED FORM OF trates the pole-zero plot for the Gp (s), which is obtained
PLANT TRANSFER from the MATLAB command, pzmap (Gp (s)). Figure 8
FUNCTION illustrates the bode plot obtained from the hardware
using the network analyzer. Figure 9 illustrates the
 s   s  bode plot obtained using MATLAB.
G po  1 + ----------- - ×  ---------------- – 1
 ω esr  ω RHP  As illustrated in Figure 8 and Figure 9, the bode plot
G p ( s ) = ------------------------------------------------------------------------------------------------------------------------ captured, using the network analyzer, matches the
 2   2 
 --------- s
s - + -----------------------  ×  --------- s
s - + -----------------------  analytical bode plot obtained in MATLAB, thereby,
- + 1 - + 1
 2 Q 1 × ω p1   2 Q 2 × ω p2  confirming the veracity of the mathematical model.
 p1
ω   p2 ω 

 2012 Microchip Technology Inc. DS01477A-page 19

AN1477
FIGURE 7: POLE-ZERO MAP OF PLANT TRANSFER FUNCTION

FIGURE 8: MEASURED BODE DIAGRAM OF PLANT TRANSFER FUNCTION

DS01477A-page 20  2012 Microchip Technology Inc.

 AN1477
FIGURE 9: SIMULATED BODE DIAGRAM OF PLANT TRANSFER FUNCTION

Digital Compensator Design for LLC In this method, an analog compensator is first designed
Resonant Converter in the continuous time domain and then converted to
discrete time domain using bilinear or tustin transfor-
The plant model is derived as shown in Equation 45. In mation. Figure 10 illustrates the block diagram of the
order to attain the desired gain margin, phase margin LLC resonant converter with a digital compensator.
and crossover frequency, a digital 3P3Z compensator
is designed. The digital 3P3Z compensator is derived
using the design by emulation or digital redesign
approach.

FIGURE 10: BLOCK DIAGRAM OF LLC RESONANT CONVERTER

Resonant Converter
VREF [n] + e[n] 3P3Z Vc[n] F(t) Vo[t]
DPWM
Compensator
–

A/D Voltage Sensor

Vm[n]
KA/D G PFc

As seen from Equation 46, plant transfer function con- Furthermore, in order to compensate for the effect of
sists of an ESR zero and a pair of dominant complex the complex dominant poles (reduction in system
poles. In order to compensate for the effect of ESR zero damping, and hence, increased overshoots and set-
(increased high-frequency gain, and thereby, tling time), two zeros, (s+a+jb) and (s+a-jb), are added.
increased ripple), a pole (ωp) is included in the compen- Also, to achieve sufficient attenuation at switching
sator. In order to minimize the steady-state error, an frequency, a pole is added to the compensator at half
integrator (Kc) is also added to the compensator. the switching frequency.

 2012 Microchip Technology Inc. DS01477A-page 21

AN1477
Effectively, the system will have a 3-Pole 2-Zero (3P2Z) If the desired crossover frequency is denoted as (fc),
compensator in continuous domain, as shown in then ωc = j2πfc.
Equation 47. At crossover frequency, the loop gain of the system
should be 0 d B or one on linear scale, as shown in
EQUATION 47: COMPENSATOR GC(s) IN Equation 48.
CONTINUOUS TIME DOMAIN
EQUATION 48: COMPENSATOR GAIN
 2  CALCULATION
s - + 1
s - + -------------------
K c ×  ------ 
2 Qc × ωz Gp ( s )
× Gc ( s ) = 1
 ωz  s = ωc s = ωc
G c ( s ) = --------------------------------------------------------------
 s   s  The required gain of the compensator is:
s ×  ------ - + 1 ×  --------- - + 1
 ωp   ω pc  1 1
K c = --------------------------------- × ---------------------------------
Gp ( s ) Gc ( s )
2
K c ⁄ ω z × ( s + α + jβ ) × ( s + α – jβ ) s = ωc s = ωc
G c ( s ) = ----------------------------------------------------------------------------------------
 s   s 
s ×  ------ - + 1 ×  --------- - + 1 The compensator first pole (ωp) is placed at 37k radi-
 ωp   ω pc  ans/second, the second pole is placed at 100k radians/
second and the complex pair of zeros is placed at 30k
radians/second. The resulting compensator for a
One of the digital compensator poles (ωp = 2πfp) is crossover frequency of 2000 Hz is shown in
placed at fp to cancel the ESR zero due to output filter Equation 49.
capacitor ESR (fesr = ωesr/2π). The compensator sec-
ond pole (ωpc) is placed at half the switching frequency EQUATION 49: COMPENSATOR TRANSFER
(fs) to obtain sufficient attenuation at the switching FUNCTION
frequency. Therefore, ωpc = 2πfs/2.
2 8
Kc represents the integral gain of the compensator and 371249.6041 × ( s + 973.6s + 8.949 × 10 )
G c = ----------------------------------------------------------------------------------------------------------
is adjusted to achieve the desired crossover frequency 4 6
s × ( s + 3.314 × 10 ) × ( s + 1.03 × 10 )
of the system.
A pair of complex zeros of the compensator, on the
complex s-plane, is at s1 = – α + jβ and s2 = – α – jβ. The
compensator zero frequency magnitude (ωz) is 2πfz.
The frequency (fz) is chosen slightly below or equal to
the corner frequency of the dominant resonant
poles (ωp1) to provide the necessary phase lead. The
compensator quality factor (Qc) is chosen to be compa-
rable or equal to the Q1 of the dominant complex pole
pair, of the plant transfer function, at the maximum load
current. In this analysis, the computation delay is
assumed to be unity.

DS01477A-page 22  2012 Microchip Technology Inc.

 AN1477
The [p, z] = pzmap (Gc (s)) command gives the poles bode plot (loop gain) obtained using the network
and zeros of the compensator. Figure 11 through analyzer, and the bode plot (loop gain) obtained using
Figure 13 illustrate the pole-zero plot for a Gc, practical MATLAB.

FIGURE 11: POLE-ZERO MAP OF COMPENSATOR

FIGURE 12: SIMULATION BODE DIAGRAM OF LOOP GAIN

 2012 Microchip Technology Inc. DS01477A-page 23

AN1477
FIGURE 13: MEASURED LOOP GAIN

The discrete compensator transfer function (Gc_d) is EQUATION 50: COMPENSATOR TRANSFER
obtained using the tustin or bilinear transformation with FUNCTION IN DISCRETE
a sampling frequency of 50 kHz, as shown in DOMAIN
Equation 50.
3 2
0.2711 × z – 0.178 × z – 0.1828 × z + 0.2663
Gc_d = -----------------------------------------------------------------------------------------------------------------
3 2
z – 0.6791 × z – 0.7342 × z + 0.4133

DS01477A-page 24  2012 Microchip Technology Inc.

 AN1477
CONCLUSION LIST OF PARAMETERS
Pulse Frequency Modulated LLC resonant converter
plant transfer function is derived by employing the EDF. TABLE 1: LIST OF PARAMETERS AND
A digital compensator is designed to meet the specifi- DESCRIPTION
cations of phase margin, gain margin and bandwidth Parameter Description
for the control system. The hardware results or wave-
forms are in conformity to the developed analytical Ir Resonant tank current
model and also meet the target specifications. Vc Resonant tank capacitor voltage
Im Magnetizing current
REFERENCES Vcf Output capacitor voltage
• “Topology Investigation for Front End DC/DC v’cf Reflected output capacitor voltage
Power Conversion for Distributed Power on primary side
Systems”, by Bo Yang, Dissertation, Virginia Isp Transformer secondary current
Polytechnic Institute and State University, 2003. Iis Sine component of resonant tank
• “Small-Signal Analysis for LLC Resonant current
Converter”, by Bo Yang and F.C. Lee, CPES Iic Cosine component of resonant tank
Seminar, 2003, S7.3, Pages: 144-149. current
• “Small-Signal Modeling of Series and Parallel
Vcs Sine component of resonant tank
Resonant Converters”, by Yang, E.X.; Lee, F.C.;
capacitor voltage
Jovanovich, M.M., Applied Power Electronics
Conference and Exposition, 1992. APEC' 92. Vcc Cosine component of resonant tank
Conference Proceedings 1992, Seventh Annual, capacitor voltage
1992, Page(s): 785-792. Ims Sine component of magnetizing
• “Approximate Small-Signal Analysis of the Series current
and the Parallel Resonant Converters”, by Imc Cosine component of magnetizing
Vorperian, V., Power Electronics, IEEE current
Transactions on, Vol. 4, Issue 1, January 1989, Iss Sine component of transformer
Page(s): 15-24. secondary current
• “DC/DC LLC Reference Design Using the
Isc Cosine component of transformer
dsPIC ® DSC” (AN1336)
secondary current
Ves Sine component of half-bridge
inverter output voltage
Vec Cosine component of half-bridge
inverter output voltage
Ips Sine component of transformer
primary current
Ipc Cosine component of transformer
primary current
Ipp Total primary current of transformer
Vps Sine component of transformer
primary voltage
Vpc Cosine component of transformer
primary voltage
n Transformer turns ratio

 2012 Microchip Technology Inc. DS01477A-page 25

AN1477
NOTES:

DS01477A-page 26  2012 Microchip Technology Inc.

Note the following details of the code protection feature on Microchip devices:
• Microchip products meet the specification contained in their particular Microchip Data Sheet.

• Microchip believes that its family of products is one of the most secure families of its kind on the market today, when used in the
intended manner and under normal conditions.

• There are dishonest and possibly illegal methods used to breach the code protection feature. All of these methods, to our
knowledge, require using the Microchip products in a manner outside the operating specifications contained in Microchip’s Data
Sheets. Most likely, the person doing so is engaged in theft of intellectual property.

• Microchip is willing to work with the customer who is concerned about the integrity of their code.

• Neither Microchip nor any other semiconductor manufacturer can guarantee the security of their code. Code protection does not
mean that we are guaranteeing the product as “unbreakable.”

Code protection is constantly evolving. We at Microchip are committed to continuously improving the code protection features of our
products. Attempts to break Microchip’s code protection feature may be a violation of the Digital Millennium Copyright Act. If such acts
allow unauthorized access to your software or other copyrighted work, you may have a right to sue for relief under that Act.

Information contained in this publication regarding device Trademarks

applications and the like is provided only for your convenience
The Microchip name and logo, the Microchip logo, dsPIC,
and may be superseded by updates. It is your responsibility to FlashFlex, KEELOQ, KEELOQ logo, MPLAB, PIC, PICmicro,
ensure that your application meets with your specifications.
PICSTART, PIC32 logo, rfPIC, SST, SST Logo, SuperFlash
MICROCHIP MAKES NO REPRESENTATIONS OR
and UNI/O are registered trademarks of Microchip Technology
WARRANTIES OF ANY KIND WHETHER EXPRESS OR Incorporated in the U.S.A. and other countries.
IMPLIED, WRITTEN OR ORAL, STATUTORY OR
OTHERWISE, RELATED TO THE INFORMATION, FilterLab, Hampshire, HI-TECH C, Linear Active Thermistor,
INCLUDING BUT NOT LIMITED TO ITS CONDITION, MTP, SEEVAL and The Embedded Control Solutions
QUALITY, PERFORMANCE, MERCHANTABILITY OR Company are registered trademarks of Microchip Technology
FITNESS FOR PURPOSE. Microchip disclaims all liability Incorporated in the U.S.A.
arising from this information and its use. Use of Microchip Silicon Storage Technology is a registered trademark of
devices in life support and/or safety applications is entirely at Microchip Technology Inc. in other countries.
the buyer’s risk, and the buyer agrees to defend, indemnify and
Analog-for-the-Digital Age, Application Maestro, BodyCom,
hold harmless Microchip from any and all damages, claims,
suits, or expenses resulting from such use. No licenses are chipKIT, chipKIT logo, CodeGuard, dsPICDEM,
dsPICDEM.net, dsPICworks, dsSPEAK, ECAN,
conveyed, implicitly or otherwise, under any Microchip
ECONOMONITOR, FanSense, HI-TIDE, In-Circuit Serial
intellectual property rights.
Programming, ICSP, Mindi, MiWi, MPASM, MPF, MPLAB
Certified logo, MPLIB, MPLINK, mTouch, Omniscient Code
Generation, PICC, PICC-18, PICDEM, PICDEM.net, PICkit,
PICtail, REAL ICE, rfLAB, Select Mode, SQI, Serial Quad I/O,
Total Endurance, TSHARC, UniWinDriver, WiperLock, ZENA
and Z-Scale are trademarks of Microchip Technology
Incorporated in the U.S.A. and other countries.
SQTP is a service mark of Microchip Technology Incorporated
in the U.S.A.
GestIC and ULPP are registered trademarks of Microchip
Technology Germany II GmbH & Co. & KG, a subsidiary of
Microchip Technology Inc., in other countries.
All other trademarks mentioned herein are property of their
respective companies.
© 2012, Microchip Technology Incorporated, Printed in the
U.S.A., All Rights Reserved.
Printed on recycled paper.
ISBN: 978-1-62076-690-3

QUALITY MANAGEMENT SYSTEM Microchip received ISO/TS-16949:2009 certification for its worldwide
headquarters, design and wafer fabrication facilities in Chandler and
CERTIFIED BY DNV Tempe, Arizona; Gresham, Oregon and design centers in California
and India. The Company’s quality system processes and procedures

== ISO/TS 16949 ==
are for its PIC® MCUs and dsPIC® DSCs, KEELOQ® code hopping
devices, Serial EEPROMs, microperipherals, nonvolatile memory and
analog products. In addition, Microchip’s quality system for the design
and manufacture of development systems is ISO 9001:2000 certified.

 2012 Microchip Technology Inc. DS01477A-page 27

 Worldwide Sales and Service
AMERICAS ASIA/PACIFIC ASIA/PACIFIC EUROPE
Corporate Office Asia Pacific Office India - Bangalore Austria - Wels
2355 West Chandler Blvd. Suites 3707-14, 37th Floor Tel: 91-80-3090-4444 Tel: 43-7242-2244-39
Chandler, AZ 85224-6199 Tower 6, The Gateway Fax: 91-80-3090-4123 Fax: 43-7242-2244-393
Tel: 480-792-7200 Harbour City, Kowloon India - New Delhi Denmark - Copenhagen
Fax: 480-792-7277 Hong Kong Tel: 45-4450-2828
Tel: 91-11-4160-8631
Technical Support: Tel: 852-2401-1200 Fax: 45-4485-2829
Fax: 91-11-4160-8632
http://www.microchip.com/
Fax: 852-2401-3431 France - Paris
support India - Pune
Australia - Sydney Tel: 91-20-2566-1512 Tel: 33-1-69-53-63-20
Web Address:
Tel: 61-2-9868-6733 Fax: 91-20-2566-1513 Fax: 33-1-69-30-90-79
www.microchip.com
Fax: 61-2-9868-6755 Germany - Munich
Atlanta Japan - Osaka
China - Beijing Tel: 81-66-152-7160 Tel: 49-89-627-144-0
Duluth, GA
Tel: 86-10-8569-7000 Fax: 49-89-627-144-44
Tel: 678-957-9614 Fax: 81-66-152-9310
Fax: 86-10-8528-2104 Italy - Milan
Fax: 678-957-1455 Japan - Yokohama
China - Chengdu Tel: 39-0331-742611
Boston Tel: 81-45-471- 6166
Tel: 86-28-8665-5511 Fax: 39-0331-466781
Westborough, MA Fax: 81-45-471-6122
Fax: 86-28-8665-7889 Netherlands - Drunen
Tel: 774-760-0087 Korea - Daegu
Fax: 774-760-0088 China - Chongqing Tel: 31-416-690399
Tel: 82-53-744-4301
Tel: 86-23-8980-9588 Fax: 82-53-744-4302 Fax: 31-416-690340
Chicago
Itasca, IL Fax: 86-23-8980-9500 Spain - Madrid
Korea - Seoul
Tel: 630-285-0071 China - Hangzhou Tel: 34-91-708-08-90
Tel: 82-2-554-7200
Fax: 630-285-0075 Tel: 86-571-2819-3187 Fax: 34-91-708-08-91
Fax: 82-2-558-5932 or
Cleveland Fax: 86-571-2819-3189 82-2-558-5934 UK - Wokingham
Independence, OH China - Hong Kong SAR Tel: 44-118-921-5869
Malaysia - Kuala Lumpur
Tel: 216-447-0464 Tel: 852-2401-1200 Fax: 44-118-921-5820
Tel: 60-3-6201-9857
Fax: 216-447-0643 Fax: 852-2401-3431 Fax: 60-3-6201-9859
Dallas China - Nanjing Malaysia - Penang
Addison, TX Tel: 86-25-8473-2460 Tel: 60-4-227-8870
Tel: 972-818-7423 Fax: 86-25-8473-2470 Fax: 60-4-227-4068
Fax: 972-818-2924
China - Qingdao Philippines - Manila
Detroit Tel: 86-532-8502-7355
Farmington Hills, MI Tel: 63-2-634-9065
Fax: 86-532-8502-7205 Fax: 63-2-634-9069
Tel: 248-538-2250
Fax: 248-538-2260 China - Shanghai Singapore
Tel: 86-21-5407-5533 Tel: 65-6334-8870
Indianapolis Fax: 86-21-5407-5066
Noblesville, IN Fax: 65-6334-8850
Tel: 317-773-8323 China - Shenyang Taiwan - Hsin Chu
Fax: 317-773-5453 Tel: 86-24-2334-2829 Tel: 886-3-5778-366
Fax: 86-24-2334-2393 Fax: 886-3-5770-955
Los Angeles
Mission Viejo, CA China - Shenzhen Taiwan - Kaohsiung
Tel: 949-462-9523 Tel: 86-755-8203-2660 Tel: 886-7-213-7828
Fax: 949-462-9608 Fax: 86-755-8203-1760 Fax: 886-7-330-9305
Santa Clara China - Wuhan Taiwan - Taipei
Santa Clara, CA Tel: 86-27-5980-5300 Tel: 886-2-2508-8600
Tel: 408-961-6444 Fax: 86-27-5980-5118 Fax: 886-2-2508-0102
Fax: 408-961-6445 China - Xian Thailand - Bangkok
Toronto Tel: 86-29-8833-7252 Tel: 66-2-694-1351
Mississauga, Ontario, Fax: 86-29-8833-7256 Fax: 66-2-694-1350
Canada China - Xiamen
Tel: 905-673-0699 Tel: 86-592-2388138
Fax: 905-673-6509 Fax: 86-592-2388130
China - Zhuhai
Tel: 86-756-3210040
10/26/12
Fax: 86-756-3210049

DS01477A-page 28  2012 Microchip Technology Inc.