AN_TDA4863(G)(-2)

A p p l i c a t i on N o t e , V 1 . 2, O c t . 20 0 3
TDA 4 863 Technical Description
AN-PFC-TDA 4863-1
Author: W. Frank
http://www.infineon.com/pfc
Power Management & Supply
N e v e r
s t o p
t h i n k i n g .
TDA 4863 - Technical Description
Revision History:
29.10.2003
Previous Version:
1.1
Page
Subjects (major changes since last revision)
2
updated
V1.2
For questions on technology, delivery and prices please contact the Infineon
Technologies Offices in Germany or the Infineon Technologies Companies and
Representatives worldwide: see our webpage at http://www.infineon.com.
CoolMOSTM, CoolSETTM are a trademarks of Infineon Technologies AG.
Edition 29.10.2003
Published by Infineon Technologies AG,
St.-Martin-Strasse 53,
81669 München, Germany
© Infineon Technologies AG 2002.
All Rights Reserved.
Attention please!
The information herein is given to describe certain components and shall not be considered as warranted
characteristics.
Terms of delivery and rights to technical change reserved.
We hereby disclaim any and all warranties, including but not limited to warranties of non-infringement, regarding
circuits, descriptions and charts stated herein.
Infineon Technologies is an approved CECC manufacturer.
Information
For further information on technology, delivery terms and conditions and prices please contact your nearest
Infineon Technologies Office in Germany or our Infineon Technologies Representatives worldwide.
Warnings
Due to technical requirements components may contain dangerous substances. For information on the types in
question please contact your nearest Infineon Technologies Office.
Infineon Technologies Components may only be used in life-support devices or systems with the express written
approval of Infineon Technologies, if a failure of such components can reasonably be expected to cause the failure
of that life-support device or system, or to affect the safety or effectiveness of that device or system. Life-support
devices or systems are intended to be implanted in the human body, or to support and/or maintain and sustain
and/or protect human life. If they fail, it is reasonable to assume that the health of the user or other persons may
be endangered.
TDA 4863 - Technical Description
Table of Contents
Page
1
Short Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2
2.1
2.2
2.2.1
2.2.2
2.2.3
2.2.4
2.2.5
2.2.6
2.2.7
Technical Description TDA 4863 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Control Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Self-Start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Driver Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Control Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Setting and Limitation of Output Voltage . . . . . . . . . . . . . . . . . . . . . . . . . 8
Multiplier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Current Comparator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Detector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3
3.1
3.2
3.2.1
3.2.2
3.3
3.4
3.4.1
3.4.2
3.4.3
Applications of the TDA 4863 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IC Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Opamp Compensation Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PI-Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PIT1 Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Universal Preconverter and 2-lamp Dimming Ballast Design . . . . . . . . . .
Design Steps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Input and Output Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Multiplier Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Boost Inductor Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
Summary of Used Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Application Note
3
12
12
13
14
15
16
17
17
19
19
V1.2, 29.10.2003
TDA 4863 - Technical Description
Short Description
1
Short Description
The TDA 4863 integrated circuit is the successor IC of TDA 4862. The main
improvements are the reduction of the startup current, a reliable gate voltage level at
worst case conditions and an improved low-load behavior. Like its predecessor it is able
to control a variety of converter topologies which are suited for power factor correction
operation. Amongst those, the boost converter is the most widespread one and is also
described in this application note. The IC controls the power switch, so that sinusoidal
current is taken from the single-phase line supply and a stabilized DC voltage is available
at the output. The circuit therefore acts as active harmonic filter and limits the
corresponding harmonic currents resulting from the capacitor pulse charge currents of
conventional rectification. The power factor which describes the ratio between active and
apparent power is almost 1. The IC also compensates possible changes of the line
voltage and provides several protection features.
Application Note
4
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
2
Technical Description TDA 4863
2.1
Control Method
The control method of the harmonic filter is based on the physical relationship between
current and voltage at the boost converter choke, i.e. uL = L diL /dt, as it is shown in
Figure 1. This leads to a triangular current waveform which is
i L ( t ) = L ⋅ Vin ⋅ t ⁄ T on ;
0 ≤ t < T on (T switched on, D switched off) [1]
i L ( t ) = i L ( T on ) – L ⋅ ( V out – V in ) ⋅ t ⁄ T off ; T on ≤ t < T on + T off (T switched off, D switched on)
while neglecting the voltage iD(t).RDS(on) of the MOSFET and VF of the boost diode,
respectively.
The transistor T does not switch on again until the current through the boost converter
diode D turns zero. This avoids reverse recovery losses in the diode and the triangles of
the inductor current iL are always set immediately one after the other which is well known
as discontinuous conduction mode (DCM). Using this control method, the pulse periods
(i.e. Tp = Ton + Toff) and the operating pulse frequency fp of the active harmonic filter
change with the input voltage and the load.
iL
vin(t)
D
vL
vD
T
|vin(t)|
vT
TDA
4863
Figure 1
L
Vout
Principle of Boost Converter Used for PFC
The mean input current calculated over a pulse period is exactly half the peak value of
the high frequency choke current. According to Figure 2, the TDA 4863 modulates them
so that they are proportional to an envelope of the sinusoidal line input voltage vin and a
sinusoidal line current iin will be drawn after smoothing by means of an RFI suppression
filter.
Application Note
5
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
V, I
vout
vin
iL
iin
t
Figure 2
Input Current Shape and Inductor Current in a PFC Boost Converter
2.2
Characteristics
2.2.1
Self-Start
The IC is supplied via resistors either from the rectified mains or from the output voltage
during start-up. In this state (VCC < 10 V typ.) the IC consumes a current of less than
100 µA (max.). An turn-on threshold of 12.5 V typically is responsible to set the IC into
an operable state. An undervoltage lockout with a turn-off threshold of 10 V typically
prevent both the IC and the boost converter to get into a dangerous state of operation.
The thresholds are also shown in Figure 3. After enabling the gate driver, a startup timer
generates a set of pulses for the turn-off flip-flop. This is also done, if the gate driver
output remains in low state levels for longer than 150 µs. In order to guarantee safe
supply of the IC, the supply voltage pin 8 is internally limited to 20 V to ground. Thus, the
IC has all functions necessary for low-loss self-start.
2.2.2
Driver Output
The driver output has been designed to control power MOSFET transistors with a current
handling capability of ±500 mA (max.). In order to avoid reverse currents the totem pole
driver output is equipped with clamping diodes connected to ground and supply voltage.
Additionally, the gate drive is clamped to a maximum value of about 11.5 V typically,
which is not shown in Figure 3.
Application Note
6
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
2.5V Reference
Undervoltage
Lockout 12,5V - 10V
Internal Supply
C2
C3
2
VOLTAGE
AMPLIFIER
OUTPUT
Restart Timer
2.2V
1
0.2V
VOLTAGE
SENSE
0.2V C4
0
&
0
&
0
0
40µA
0
0
8
0
S
Q
R
Q
5k
Figure 3
&
VCC
0
0
7
DRIVE
OUTPUT
0
6
200ns
LEB
GROUND
0.6V
M2
M3
VFB ... VFB +1.5V
M1
Multiplier
0V...4V
MULTIPLIER
IN
5V
20V
0
3
ZERO
CURRENT
DETECTOR
5
1.5V
1.0V
TDA
4863
C5
VOP
1.2V
C1
QM=M1*(M2-VFB)*K
0.3V-1 < K < 0.7V-1
VFB=2.5V
10k
4
CURRENT
SENSE
1.0V
Scheme of TDA 4863
The driver output actively controls LOW levels during standby state using a residual
voltage of 1 V at a gate current IGT = 20 mA dissipation current.
2.2.3
Control Amplifier
The control amplifier compares the divided output voltage at its inverting input with a highly
accurate reference voltage of 2.5 V at its non-inverting input. The reference has a maximum
deviation of less than ±2% over the total temperature range (-40°C < TJ < 150°C). A
feedback network is inserted between the amplifier output (pin 2) and its inverting input
(pin 1) for the purpose of frequency response compensation. Using only one capacitor as
an integral controller causes oscillating transient response, because the boost converter
acts as a controlled current source and the storage capacitor at its output delays the phase
by almost 90° in no-load and in low-load operation. A compensation network design
containing a PI- or a PIT1-controller will lead to much better results. This is also described
in the section “Design Steps” on Page 17.
The output voltage VVAOUT of the control amplifier ranges from 1.1 V to 5.4 V and can be
loaded with a current of 6 mA (source) and 30 mA (sink), respectively and is monitored
by comparator C3 according to Figure 3. If the output voltage drops 0.3 V below the
reference level of 2.5 V (i.e. reference voltage), the driver output will be blocked directly
Application Note
7
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
via the turn-off flip-flop. This measure guarantees the stability of the output voltage in
case of complete no-load operation, without interferences from offset voltages at the
multiplier output or at the comparator input.
On the other hand if the voltage sense signal drops below 0.2 V, the IC disables the gate
drive. That may occur e.g. when the feedback signal is broken. As soon as the supply
voltage crosses the UVLO level, the IC is in off-state and begins another start-up cycle.
This is not performed, if the IC is supplied from the subsequent converter, because the
supply voltage stays stable and the UVLO level is never reached. Then the IC is
continuously off.
The output voltage of the boost converter contains a superimposed AC voltage having
twice the mains frequency. Its amplitude depends on the output capacitor and the load
and is also fed back via the control amplifier. That causes undesired modulations and
increases the total harmonic distortion of the line current drawn. Therefore a bandwidth
of the control amplifier is chosen which is considerably lower than twice the mainsfrequency. However, the controller then reacts slowlier to sudden load changes which
may result in temporary overvoltages or even output breakdowns.
2.2.4
Setting and Limitation of Output Voltage
V0=445V
+
C4
-
436V
396V
VREF
Vcc
RH
748
k
I=591µA
M3
+
OP
-
M2
∆I
Voltage
Sense
∆I=0
2,5V
1
Threshold
∆I=40µA
RH
748
k
I=631µA
∆I
2,5V
=40µA
RH
1M0
I=433µA
∆I
=40µA
2,5V
I=591µA
I=591µA
I=393µA
RL
4k2
2
RL
4k2
2
RL
6k3
4
∆I
0,6V
TDA 4863
2
Figure 4
V0
475V
445V
U+
Voltage Amplifier Output
V0
Compensation
Network
Overvoltage Limitation Dependent on the Output Voltage Divider
The output voltage is set by a voltage divider from Vout to GND. The divider should be
high ohmic in order to prevent losses. The tap of the divider is set to 2.5 V (i.e. reference
voltage, see section “Design Steps” on Page 17). But voltage transients or load
Application Note
8
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
rejections may cause output voltage overshoots in boost converter which makes an
overvoltage protection necessary, e.g. for the output capacitor. The higher output
voltage also forces a higher current through the output voltage divider which causes an
increase of the tap potential. The voltage amplifier now tries to compensate this by
measuring the current ∆I flowing into the compensation network according to Figure 5.
∆I is therefore dependent on the inner resistance of the divider. So is the overvoltage
protection limit VOVP because as soon as ∆I exceeds 40 µA the comparator C4 in
Figure 5 is effective via multiplier input M3. Then the multiplier output is almost linearly
reduced until it reaches about 0.1 V at 43 µA. This is shown in diagram 7 of the
preliminary datasheet [1]. Since the multiplier output simultaneously is the reference
voltage of the current comparator its decrease also affects a decrease of the line current
drawn. This technique continuously compensates the input current back and avoids
uncontrolled oscillations of the line current drawn, as it usually appears with digital
measures.
That means, the output voltage divider has to meet the following conditions:
V out = V ref ⋅ ( R H + R L ) ⁄ R L and
[2]
V OVP = V ref + R 4 ⋅ ( ∆I + V ref ⁄ R L )
[3]
Therefore the resistor values are
V ref V OVP – V out
R L = ---------- ⋅  -------------------------------
∆I  V out + V ref 
[4]
V OVP – V ref
R H =  --------------------------------
 ∆I + V ref ⁄ R L
[5]
with the reference voltage Vref = 2.5 V
The overvoltage control is also guaranteed in those states of operation when the output
voltage of the control amplifier reaches its upper limit threshold, because the dissipation
current is measured as well. As soon as the output voltage of the control amplifier tends
towards the minimum level, the comparator turns off at a level of 2.2 V to guarantee safe
no-load operation.
2.2.5
Multiplier
The multiplier has three inputs M1, M2 and M3. The input M1 provides the information of
the waveform of the input voltage. Its range is from 0 V to 4 V being referenced to ground.
In order to reach a high THD value it is necessary that the transfer function Mout / M1 is
highly linear like it is shown in diagram 9 of the datasheet [1]. The best performance is
available, when the tap potential of the AC voltage divider is close to 4.0 V at highest
peak input voltage.
Application Note
9
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
The signal M2 is the output of the voltage amplifier with a range of 2.5 V to 4 V which is
referenced to 2.5 V. It is the compensated feedback signal of the output voltage.
Principally, the control loop gain will increase with increasing input voltages at given
output power which possibly results in control loop instability. But it also causes a high
amplification of the 100 Hz-ripple of the DC-bus voltage or other disturbances and is
therefore responsible for poor THD values. Unlike input M1, it needs not necessarily to
be highly linear. As it is already shown with other PFC control IC better results in terms
of THD are possible with a non linear transfer function like square or exponential
functions ([2], [3]). A piecewise linear transfer function is implemented in the TDA 4863
according diagram 10 of the datasheet in order to decrease the loop gain in such critical
states of operation.
The two corresponding constant factors Klow and Khigh are 0.3 V-1 and 0.7 V-1,
respectively, which are internal factors of the multiplier. Their dimension is V-1 in order to
comply with the following equations. In this way the current comparator level can be
calculated as
VQM = Klow (Vpin2 - VREF) Vpin3
for VVVAOUT < 3.0 V
VQM = Khigh (Vpin2 - VREF) Vpin3
for VVVAOUT > 3.5 V
The output voltage of the multiplier is limited to 1.0 V. This measure causes a defined
turn-off threshold for current limitation and also leads to a higher gate voltage uGS of the
power MOSFET at worst case conditions.
2.2.6
Current Comparator
The current comparator detects the momentary shunt resistor voltage via its inverting
input. The shunt resistor is in the source path of the MOSFET and should have an
intrinsic inductance as low as possible. As soon as the shunt voltage reaches the turnoff threshold defined by the multiplier, the turn-off flip-flop is reset and the driver switches
off. The turn-off flip-flop prevents multiple pulses during the switching procedure of the
power MOSFET.
Voltage spikes occur at the shunt resistor during the turn-on of the MOSFET as a result
of the intrinsic inductance of the shunt and the influence of the driver currents. They may
lead to an irregular turn-off of the MOSFET. An integrated leading edge blanking
suppresses such effects by making the current comparator signal ineffective for 200 ns
(typ.).
2.2.7
Detector
A second winding (detector winding) on the choke provides an image of the drain voltage
in the ratio Ndet /Nboost according to Figure 6. During interval T1 the MOSFET is on
resulting in a negative voltage at the detector winding. When the MOSFET turns off, the
flip-flop is reset by the current comparator. Simultaneously the polarity of the detector
winding will change to positive corresponding to the drain voltage. When the detector
Application Note
10
V1.2, 29.10.2003
TDA 4863 - Technical Description
Technical Description TDA 4863
voltage exceeds 1.5 V with positive edge the flip-flop is locked. After the diode took the
current its voltage is positive (interval T2) as long as current is flowing through the boost
diode. When the current reaches zero, the detector voltage falls below 1 V being the level
of enabling the driver stage for the MOSFET.
The calculation of its number of turns is done in section “IC Supply” on Page 12. The
winding is applied to pin 5 via a high-ohmic resistance (10 kΩ to 47 kΩ). Clamping
structures are available in the IC which limit the voltage at the input to +5 V and +0.5 V,
respectively, at ±5 mA.
There are cases in which there is no significant detector signal to set the turn-off flip-flop.
This may occur when the supply voltage is switched on, in case of line overvoltage
exceeding the output voltage and in no-load and low-load operation, when the voltage
controller specifies intermittant operation. In that case a startup generator is activated
which supplies a set of pulses to the turn off flip-flop if the driver output stays on LOWlevel longer than 160 µs typically.
T1
Figure 5
T2
Drain voltage (black, 100 V/div), detector voltage (red, 10 V/div) and
shunt voltage (200 mV/div) using a supply circuit similar to Figure 10
Application Note
11
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
3
Applications of the TDA 4863
3.1
IC Supply
An obvious way to supply the IC is to use the detector winding. We have to care, that the
supply circuit doesn't influence the detector signal. First, in a simple voltage mode supply,
we use a diode, a storage capacitor C10 and a current limiting resistor R12 as it is given in
a) of Figure 6. We achieve good results in ballast applications with the following design
of the transformation ratio:
V ZCD
N ZCD
-------------- = ---------------------------------NP
V out – V innom
VZCD = 22 V...24 V;
R12 = 220 Ω...270 Ω
[6]
Second, in a charge pump supply we use two diodes, two capacitors C10, C13 and one
decoupling resistor R12 or a decoupling inductor L5 (lower losses) and a current limiting
resistor R12A as it is shown in b) and c) of Figure 6, to avoid burn down at resonance
frequency. This method of supply is to prefer in SMPS applications with wide input
voltage range. All the following applications use circuitry b) of Figure 6.
The supply current increases with the operating frequency at low load and is not
dependent on the input voltage. We achieve good results with the following design of the
transformation ratio:
V ZCD
N ZCD
------------- = ------------NP
V out
VZCD ≈ 80 V,
C13 = 3 nF...4 nF;
R12 = 390 Ω...270 Ω
[7]
Or C13 = 1 nF...1.5 nF, L5 = 50 µH...100 µH, R12A designed with C13 and L5 as a low-pass
filter of Bessel characteristic.
Application Note
12
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
NP
a)
NZCD
C10
R12
D6
ZCD
Vcc
b)
NP
NZCD
D7
C10
D6
R12
C13
NP
c)
NZCD
D7
R12A
D6
C13
Vcc
Figure 6
3.2
R9
ZCD
Vcc
C10
R9
L5
R9
ZCD
IC Supply Circuit Realized with Rectifier (a) and Charge Pump (b and c)
Opamp Compensation Design
The design of the compensation network of PFC controller is a very sensitive topic,
because it highly influences the performance of the circuit in terms of the total harmonic
distorsion (THD) of the input current. It is particularly responsible for the control loop
behavior in case of changes of operational parameters, such as load or input voltage, as
well as for the AC-ripple rejection of the output voltage. Especially the demand for a good
suppression of the superimposed AC-share of the output voltage is contrary to the
demand of a good behavior at load changes or changes of the input voltage, because
they take effect in different areas of frequencies. For example, load steps will always
affect higher frequencies than the 100 Hz (or 120 Hz in North America) output voltage
ripple does.
Application Note
13
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
3.2.1
PI-Controller
In case that a constant output power is taken from the boost converter, the boost
converter can be modelled as an integrator according to [2]
2V in
KI = --------------4V out
V out
KI
F S ( s ) = -----------------------= -------------;
I PkmaxHF
sC out
a)
RY
Vin,OP
b)
R2
Figure 7
Vin,OP
=
1+jf/f2
jf/f1
C1 = 1/(2π*f1*RY)
R2 = 1/(2π*f2*C1)
GP = 20 log(f1/f2)
|F|res
|F|C
f2
fC f1
0
-180°
VO,OP
|F|S
-20
-90°
FC =
VO,OP
2,5 V
40
20
GP
0
C1
[8]
FS
FC
Fres
f
Φr
f
fC
a) PI-Controller Topology
b) Frequency Response of PI-Controller and Load
It has therefore a phase of constantly 90°. If the compensation would now be integrating
as well, the resulting phase response of the control-loop will have 180°, which will lead
to a phase margin Φr = 0 and to a highly instable system. Hence, a PI-controller topology
is recommended at least. Such a controller is shown in a) of Figure 8
The correspondence between the components RY, C1 and R2 is given by the equations
in Figure 7.
Part b) of Figure 7 shows the resulting frequency response of the transfer function of the
controller of a) and a load with constant power. In the area of the crossover frequency fC
Application Note
14
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
there is a decrease of 20 dB/decade which indicates a good dynamic behavior of the
system [4]. There is also a zero in the Nominator of the controller transfer function giving
a phase shift of +90°. In Figure 7 this leads to a phase margin of Φr = 90° and therefore
a good stability of the system.
A good set of the frequencies f1 and f2 will be f1 = 0.1 Hz and f2 = 5 Hz which will result
in a capacitor C1 = 2.2 µF and a resistor R2 = 16 kΩ.
It is usually recommended, that the bandwidth of the compensation is about 20 Hz. This
will lead to a relatively constant output of the compensator over a mains period. A better
performance can be achieved by introducing a pole at 100 Hz (or 120 Hz, respectively).
C2
a)
RY
Vin,OP
b)
R2 C1
VO,OP
2,5 V
20
GP
0
-180°
Figure 8
Vin,OP
=
1+jf/f2
jf/f1 (1+jf/f3)
C1 = 1/(2π*f1*RY)
C2 = C1/((f3*/ f2 )-1)
R2 = 1/(2π*f2*C1)
GP = 20 log(f1/f2)
|Fres|
|FC|
f
f2
f4
fC f1 f3
0
-90°
VO,OP
|FS|
40
-20
FC =
FS
FC
Fres
Φr
f
fC
a) PI-Controller topology
b) Frequency Response of PIT1-Controller and Constant Load
Application Note
15
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
3.2.2
PIT1 Controller
In some applications a higher immunity against higher frequency disturbances is
required. This can be achieved using an additional delay share in the controller transfer
function as it is shown in Figure 8 causing a pole in the controller transfer function. The
only difference is the additional capacitor C2 in parallel to the resistor R2 as it is shown in
a) of Figure 8.
The resulting gain response |Fres| now decreases with 40 dB/decade after frequency f3,
so that higher frequencies will be damped stronglier. Simultaneously the trace rises with
40 dB/decade on direction of lower frequencies which will amplify low frequency shares
of the opamp input voltage. Both will lead to a better steady state operation and the
influence of disturbances will be reduced. It must be noted here, that the phase margin
must not necessarily be 90° as it is shown in b) of Figure 8. With another selection of f3
there might also be smaller phase margins.
A recommended rating for a 110 W PFC preconverter for wide range applications is
C1 = 2.2 µF, C2 = 1 µF and R2 = 33 kΩ.
3.3
Universal Preconverter and 2-lamp Dimming Ballast Design
These applications demonstrate the excellent performance of the TDA 4863 controlling
a power factor preconverter. The design steps indicate the method of the calculation of
the components values. A design for lamp ballast (nominal input voltage Vinnom = 277 V,
output power Pout = 70 W) as well as a design for switched mode power supplies (input
voltage Vin = 90 V ... 265 V, output power Pout = 120 W) are give here as examples.
Circuit diagrams and measurements results at different operating conditions establish a
good basis for evaluation.
Usually a single stage RFI-filter does not accomplish the RFI-standards. Therefore
multiple stage RFI-filters are designed into these applications as an example how to
suppress resonant oscillations of these filters.
Discontinuous conduction mode always results in high efficiency, because it avoids
reverse recovery losses of the boost converter diode. A high power factor, low
harmonics, a wide input voltage range and a feedback controlled output voltage are the
most important features of a power factor preconverter. The TDA 4863 contains all
control and monitoring functions to meet these demands.
Application Note
16
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
3.4
Design Steps
3.4.1
Input and Output Section
Application
Nominal input voltage
Minimum input voltage
Maximum input voltage
Minimum peak input voltage
Maximum peak input voltage
Estimated minimum efficiency
Output power
Maximum peak input current
Maximum high frequency peak current
Maximum current sense threshold
Shunt resistor
Nominal output voltage
Reference voltage
Controller current at pin VAOUT
Overvoltage threshold
Output voltage divider
Vinnom
Vinmin
Vinmax
VinPkmin
VinPkmax
η
Pout
IinPkmax
IPkmaxHF
VISensemax
R11
Vout
Vref
IVAOUT
VOV
R4
R5
Application Note
2-Lamp-Ballast SMPSPreconverter
277 V AC
110 V - 230 V
250 V AC
90 V AC
305 V AC
265 V AC
353 V
127 V
431 V
375 V
0.9
0.9
70 W
120 W
0.44 A
2.1 A
0.88 A
4.2 A
= Vinnom - 20%
= Vinnom + 20%
= √2 Vinnin
= √2 Vinnax
= η Pin
= 2Pout /(Vinpmin η)
= 2IinPkmax
= 1.0 V
= VISensemax /IPkmaxHF
1.14 Ω
Recommended minimum: Vout +30 V 450 V DC
= 2.5 V
= 40 µA
≈ 1.1 Vout
480 V DC
748 kΩ
Vref VOVP − Vout
= R5 =
)
⋅(
∆I
=
R4 = (
17
0.24 Ω
400 V DC
440 V DC
998 kΩ
Vout + Vref
VOVP − Vref
∆I + Vref / R5
4.12 kΩ
6.34 kΩ
)
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
3.4.2
Multiplier Section
Application
1-Lamp-Ballast SMPSPreconverter
Multiplier inputs M1 and M2 dynamic
voltage range
Vm1R = 4.0 V; Vm2R = 1.5 V
Multiplier output limitation
VQmmax
= VISensemax = 1.0 V
Multiplier gain
Vm1(VQm = 1.3 V; Vm2R = 2 V)
Select Vm1 = Vm1lim = 1.2 V
K
= 0.65
Select upper resistor of input voltage divider
R6
Lower resistor of input voltage divider
R7
Low pass filter capacitor
Test: Input range
Vm1(Vin =Vinpmax) < Vm1R = 3.8 V?
otherwise select
Vm1(Vin =Vinpmin) < Vm1lim = 1.2 V
3.4.3
@Vinpmin =
127 V
1.0 MΩ
940 kΩ
= R6 . Vm1lim /(Vinpmin -Vm1lim)
9.1 kΩ
9.1 kΩ
C4
= 1/(2π.R7.f), 1 kHz < f < 3 kHz
10 nF
10 nF
Vm1inmax
= Vinpmax *R7 /(R6 + R7)
3.9 V
3.59 V
Vm1inmin
= Vinpmin *R7 /(R6 + R7)
3.2 V
1.22 V
Boost Inductor Section
In this section two different approaches for the calculation of the transformer primary inductance LP are presented. The first one is
recommended for a small input voltage range application or for applications with nearly constant output power, e.g. lamp ballasts.
The other one is suitable for the demands of wide range applications like they occur in SMPS. All the values of the sections before
are still valid.
Application Note
18
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
2-Lamp-Ballast
SMPS-Preconverter
On-time of power switch: Ton = LP . IpmaxHF /Vin, IpmaxHF = 2 Iin
Off-time of power switch: Toff = LP . IpmaxHF /(Vout - Vin)
Pulse frequency:
1
Vin ⋅ (Vout − Vin )
=
f =
Ton + Toff
Vout ⋅ LP ⋅ I p max HF
Design Criterion:
Calculate LP according to desired range of pulse frequency
(e.g. 80 kHz < fp < 110 kHz) at nominal input voltage Vinnom and
rated output power Pout
LP
=
=
Vinnom ⋅ (Vout − Vinnom )
Vout ⋅ f p ⋅ I p max HF
=
Vinnom ⋅ (Vout − Vinnom ) ⋅ η ⋅ Vinnom
Vout ⋅ f p ⋅ 2 Pout
277 V ⋅ ( 480 V − 277 V ) ⋅ 0,9 ⋅ 277 V
450 V ⋅ 90 kHz ⋅ 2 ⋅ 70 W
Calculate LP in order to obtain pulse frequencies higher than
25 kHz at maximum peak input voltage and twice of nominal
output power and at minimum peak input voltage and twice of
nominal output power
=
LP
= 1077 µH
<
2
Vinp
max ⋅ (Vout − Vinp max ) ⋅ η
Vout ⋅ f p ⋅ 2 ⋅ 2 Pout
=
(375V ) 2 ⋅ (410 V − 375 V ) ⋅ 0,9
= 665 µH
410 V ⋅ 25 kHz ⋅ 2 ⋅ 2 ⋅ 120 W
=
(127V ) 2 ⋅ (410 V − 127 V ) ⋅ 0,9
= 828 µH
410 V ⋅ 25 kHz ⋅ 2 ⋅ 2 ⋅ 120 W
and
Also possible
Calculate LP by selecting the on-time Ton in the range of
3 µs < Ton < 6 µs
LP
T ⋅V
= on innom
I p max HF
=
2
Ton ⋅ V innom
⋅η
2 ⋅ Pout
=
3 µs ⋅ (277V 2 ) ⋅ 0,9
2 ⋅ 70 W
= 1480 µH
LP
<
2
Vinp
min ⋅ (Vout − Vinp min ) ⋅ η
Vout ⋅ f p ⋅ 2 ⋅ 2 Pout
We therefore select LP < 665 µH
Core selection: E36
Both inductances are appropriate, but the calculation is done with
1077 µH. Core selection: EF20
Application Note
19
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
Number of turns:
AL-value for EF20-core:
AL
2
Bmax
⋅ Ae2
i p2 max HF ⋅ LP
=
=
AL-value for E36-core:
(250mT ) 2 ⋅ (32mm 2 ) 2
(0,88 A) 2 ⋅ 1077 µH
= 77 nH
Air gap:




1
K2
 77 nH
= 
 60,6nH



1
− 0 , 737
= K1 ⋅ s K 2
 A
s =  L
 K1nH
= 0,74mm
Select air gap according [1]: s = 0.75 mm => resulting AL-value:
=
(4,2 A) 2 ⋅ 665µH
= 77 nH




1
K2
1
 77 nH  − 0, 749
= 

 182nH 
= 3,15mm
Select air gap according [1]: s = 2 mm =>
= 76nH
AL
=>
N PFC
=
(250mT ) 2 ⋅ (120mm 2 ) 2
Air gap:
 A
s =  L
 K1nH
AL
AL
2
Bmax
⋅ Ae2
i 2p max HF ⋅ LP
= K1 ⋅ s K 2
= 108nH
=>
=
LP
AL
=
1077 µH
76nH
= 118.
N PFC
=
LP
AL
=
665µH
108nH
= 78.
Zero current detector and auxiliary power supply
A winding ratio NPFC : Naux of about 5:1 is recommended for proper Dito
operation. Therfore: Naux = 118/5 = 23
Naux = 78/5 = 16
Application Note
20
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
1077µH
L1
450V DC
MUR160
VOUT
D5
D2
2x1N4148
VIN
277V AC
+/-10%
C11
µ1
(X2)
120Ω
L1: 1077µH
E20/10/6, N27, gap=0.75mm
W 1=118 Wdg., 0.32d
W 2=23 Wdg., 0,22d
SPP4N60C2 : 600V, 0,95Ω
C6
µ1
(X2)
R14
S14
K250
C7
3n3(Y)
R8A
120k
120k
5
R10
3
C4
10n
8
D4
C9
µ15
Figure 9
R9
33k
R12
3n3/400V 470Ω
R8
R6
499k
R7
9k1
D3
D6
R6A
499k
B250C5000/3300
R15
C5
3n3(Y)
SIOV
M0.6A
0.4mH
L3
U
2x6.8mH
L2
Fuse
C13
D7
C10
47µ/25V
TDA4863
D1
6
7
1
Q1
SPP04N60
C2
12Ω
R4A
374k
R4
374k
22µF
500V
C1
2,2µ
C2
1µ
2
4
C8
R2
4k4
R11
0R47||
0R47
R5
4k22
GND
Schematic of 70 W PFC Preconverter for Two-Lamp-Ballast and Nominal Input Voltage Vin = 277 V
Application Note
21
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
Measurement Results of TDA 4863 for 2-lamp-ballast (70 W) at 10%, 25%, 50% and 100% of Rated Load
TDA 4863 Operation 70 W (Vinnom = 277; Cout = 22 µF)
Vin
[V AC]
Iin
[mA rms]
Pin
[W]
PF
THD
[%]
VBUS
[V DC]
Iout1
[mA]
Pout
[W]
Effi`cy
[%]
250.4
42.65
9.74
0.913
11.1
453.0
15.4
6.98
0.72
250.4
83.70
20.40
0.974
11.7
453.0
38.5
17.44
0.85
250.6
123.70
38.15
0.990
9.4
453.0
77.0
34.88
0.91
249.8
296.20
73.70
0.996
6.3
453.0
154.1
69.81
0.95
276.2
40.20
9.61
0.866
20.8
453.0
15.4
6.98
0.73
277.5
75.90
20.30
0.963
12.5
453.0
38.5
17.44
0.86
277.7
138.60
38.00
0.987
9.9
453.0
77.0
34.88
0.92
276.9
267.00
73.50
0.994
8.0
453.0
154.0
69.76
0.95
305.3
41.00
9.20
0.726
48.0
453.0
15.4
6.98
0.76
305.2
69.50
20.16
0.949
12.0
453.0
38.5
17.44
0.87
304.8
126.60
37.90
0.981
11.0
453.0
77.1
34.93
0.92
304
263.30
73.40
0.993
8.1
453.0
154.1
69.81
0.95
Application Note
22
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
665µH
L1
410V DC
MUR460
VOUT
D5
D2
2x1N4148
VIN
90-270V
AC
C11
µ47
(X2)
120Ω
L1: 665µH
E36/11, N27, gap=2mm
W 1=78 Wdg., 40x0.1d
W 2=16 Wdg., 0,3d
SPP4N60C2 : 600V, 0,95Ω
C6
µ47
(X2)
R14
S14
K250
C7
3n3(Y)
R8A
120k
120k
5
R10
3
C4
10n
8
D4
C9
µ15
Figure 10
R9
33k
R12
3n3/400V 470Ω
R8
R6
470k
R7
9k1
D3
D6
R6A
470k
B250C5000/3300
R15
C5
3n3(Y)
SIOV
M0.6A
0.4mH
L3
U
2x6.8mH
L2
Fuse
C13
D7
TDA4863
D1
6
7
1
Q1
SPP04N60
C2
R4A
499k
R4
499k
12Ω
100µF
450V
C1
2,2µ
C2
1µ
2
4
C10
47µ/25V
C8
R2
33k
R11
0R47||
0R47
R5
6k43
GND
Schematic of 120 W PFC Preconverter for Universal Input Voltage Range
Application Note
23
V1.2, 29.10.2003
TDA 4863 - Technical Description
Applications of the TDA 4863
Measurement results of TDA 4863 with Universal Input Voltage Range at 10%, 25%, 50% and 100% of Rated Load
TDA 4863 Operation 120 W Universal Input Voltage Range (Cout = 100 µF)
Vin
Iin
Pin
PF
THD
VBUS
Iout1
Pout
Effi`cy
[V AC]
[mA rms]
[W]
[%]
[V DC]
[mA]
[W]
[%]
90
166.10
14.80
0.987
5.9
401.0
30.0
12.03
0.81
90.2
380.10
33.30
0.997
4.2
401.0
75.0
30.08
0.90
91.5
711.00
65.00
0.998
6.0
401.0
150.0
60.15
0.93
90.6
1482.00
132.80
0.995
8.2
401.0
299.0
119.90
0.90
119.3
130.70
15.00
0.962
8.5
401.0
30.0
12.03
0.80
120.5
278.80
33.30
0.993
4.4
401.0
75.0
30.08
0.90
121
532.70
64.38
0.998
2.2
401.0
150.0
60.15
0.93
120
1069.00
128.20
0.999
4.0
401.0
299.0
119.90
0.94
180.4
100.10
15.10
0.836
12.8
401.0
30.0
12.03
0.80
181.1
191.80
33.40
0.959
7.9
401.0
74.7
29.95
0.90
182.1
355.20
64.00
0.989
4.4
401.0
150.5
60.35
0.94
181.2
693.80
125.40
0.997
3.2
401.0
300.0
120.30
0.96
231.3
102.90
14.80
0.622
37.0
401.0
30.0
12.03
0.81
230.9
156.66
33.10
0.899
9.5
401.0
74.7
29.95
0.90
230.8
284.40
63.73
0.970
6.7
401.0
150.0
60.15
0.94
230
546.00
124.80
0.993
3.4
401.0
300.5
120.50
0.97
265
102.80
14.70
0.54
26.0
401.0
30.0
12.03
0.82
265
151.30
32.80
0.817
23.8
401.0
75.0
30.08
0.92
264.7
252.60
63.40
0.948
8.8
401.0
150.0
60.15
0.95
263.9
475.70
123.90
0.987
4.6
401.0
300.0
120.30
0.97
Application Note
24
V1.2, 29.10.2003
TDA 4863 - Technical Description
Summary of Used Nomenclature
4
Summary of Used Nomenclature
Physics:
General identifiers:
Special identifiers:
A..........cross area
b, B......magnetic inductance
d, D .....duty cycle
f...........frequency
i, I........current
N .........number of turns
p, P......power
t, T.......time, time-intervals
v, V ......voltage
W.........energy
η..........efficiency
AL ........... inductance factor
V(BR)CES .. collector-emitter breakdown
voltage of IGBT
VF ........... forward voltage of diodes
Vrrm .......... maximum reverse voltage of diodes
big letters:
constant values and
time intervals
small letters: time variant values
K1, K2 ..ferrite core constants
Components:
C .........capacitance
D .........diode
IC ........integrated circuit
L..........inductance
R..........resistor
TR .......transformer
Indices:
AC.......alternating current value
DC.......direct current value
BE .......basis-emitter value
CS.......current sense value
OPTO..optocoupler value
P .........primary side value
Pk........peak value
R........... reflected from secondary to primary side
S .........secondary side value
Sh .......shunt value
UVLO ..undervoltage lockout value
Z..........zener value
Application Note
25
fmin......... value at minimum pulse frequency
i ..............running variable
in ............input value
max ........maximum value
min .........minimum value
off ...........turn-off value
on ...........turn-on value
out ..........output value
p .............pulsed
rip ...........ripple value
1, 2, 3 .....on-going designator
V1.2, 29.10.2003
TDA 4863 - Technical Description
References
5
References
[1]
Infineon Technologies AG: TDA 4863 - Power factor and boost converter
controller for high power factor and low THD; Preliminary data sheet; Infineon
Technologies AG; Munich; Germany; 07/01.
[2]
C. Adragna: Control Loop Modeling of L6561-based TM PFC; Application Note
AN 1089; STMicroelectronis; Italy; 03/01.
[3]
Infineon Technologies AG: TDA16888 - High Performance Power Combi
Controller; Preliminary Data Sheet; Infineon Technologies AG; Munich
Germiany; 02/98.
[4]
G. Schmidt: Grundlagen der Regelungstechnik; Springer Verlag; 1981.
Application Note
26
V1.2, 29.10.2003
In f i n e o n g o e s f or B u s i n e s s E x c el len c e
“Business excellence means intelligent approaches and clearly
defined processes, which are both constantly under review and
ultimately lead to good operating results.
Better operating results and business excellence mean less
idleness and wastefulness for all of us, more professional
success, more accurate information, a better overview and,
thereby, less frustration and more satisfaction.”
Dr. Ulrich Schumacher
www.infineon.com
Published by Infineon Technologies AG