A 48 V/ 2 A High Efficiency, Single Stage, Isolated Power Factor Corrected Power Supply for LED Drivers and Telecom Power

AND8394/D
A 48 V, 2 A High Efficiency,
Single Stage, Isolated
Power Factor Corrected
Power Supply for LED
Drivers and Telecom Power
http://onsemi.com
Prepared by: Frank Cathell
ON Semiconductor
Introduction
This application note describes a relatively novel and
simple way to produce an off−line, power factor corrected
15 Vdc and higher output power supply using an isolated,
single stage conversion topology. The power topology is
essentially a buck−boost derived flyback converter
operating in continuous conduction mode (CCM) and
utilizing ON Semiconductor’s NCP1652A controller which
was designed specifically for this implementation. The
power supply described in this application note is a 48 Vdc,
2 A supply with universal AC input and is intended for
general purpose supplies, telecom distributed power
“front−ends” and constant voltage LED drivers such as
those used in area lighting and distributed lighting
applications such as refrigerator case lighting and cove
lighting. Efficiencies approaching 90% were achieved with
a power factor exceeding 0.95 for most typical loads. The
supply includes overcurrent protection, overvoltage
protection, brownout detection, and an input EMI filter.
Background
Applications that require an isolated, regulated output
voltage along with input power factor correction typically
involve a two stage conversion process as depicted in
Figure 1. This scheme is composed of an input boost power
factor corrector stage which converts and pre−regulates the
input line into a 400 Vdc bus. This bus then provides the
voltage for a conventional dc−to−dc converter which can be
of any appropriate topology. For lower power applications
of 150 W and less, this is usually a flyback converter.
+
AC
input
EMI
Boost
PFC
Filter
+
Main
Converter
DC
output
Figure 1. Conventional 2−Stage Conversion
With a few minor performance compromises, a simpler
technique can be used in which the power factor and main
converter sections are combined into one conversion stage.
This has significant advantages as it eliminates the need for
the bulky boost inductor, High Voltage MOSFET, power
rectifier and bulk capacitor. This is illustrated in Figure 2.
The difference here is that the flyback conversion stage
not only handles the voltage regulation and input to output
isolation functions, but provides power factor correction as
well. The circuit essentially functions as a conventional PFC
converter with the output being derived from a secondary
winding on what would be the boost choke in the “normal”
type of non−isolated PFC circuit. The dc input to the
converter is a 100/120 Hz haversine instead of a pure dc
voltage because the normal input “bulk” capacitor following
the bridge rectifier is reduced to a value of a microfarad or
less. The full schematic of the single stage converter is
shown in Figure 3.
+
AC
input
EMI
Filter
Main
Converter
DC
output
Figure 2. Single Stage Conversion with NCP1652A
© Semiconductor Components Industries, LLC, 2010
January, 2010 − Rev. 5
1
Publication Order Number:
AND8394/D
_
0.1
C15
0.1
C27
2.2nF ”Y”
C14
10nF
MMSZ
5252B
33nF
C12
R14
470pF
8.6k
+
30.1k
2.2k
L2
R12
30.1K
0.47 ”X”
L1
C1
100k
C13
C11
C10
C9
R1
1M
0.5W
AC
In
39k
25V
4.7uF
7.32k
1nF
680pF
Notes:
1. Crossed schematic lines are not connected.
2. Heavy lines indicate power traces/planes.
3. Z2/D9 is for optional OVP (not used).
4. L1 is Coilcraft BU10−1012R2B or equivalent.
5. L2 is Coilcraft P3221−AL or equivalent.
6. L3 is Coilcraft RFB0807−3R3L or equivalent.
7. Q1 and D8 will require small heatsinks.
D9 MMSD
4148T
Z2
9
11
10
6
7
8
16
15
14 NC
U1 13
12
NCP1652
1
2
3
4
5
332k
365k
R17
R16
R15
49.9k
NCP1652A 90 W Power Supply
48 V, 2 A Out, 90−265 VAC Input (Rev 3)
0.1uF
C16
Z3
C17
MMSZ
5248B
R21
10 W
R22
10k
10 W
R25
1nF
R13
F1
2.5A
U3 TL431A
1uF
3
R20
0.1 W
0.5W
2
24k
R28
R27
1k
C26
C25
0.1
100V
R30
5.6k
R29
102k
Z4 (24V)
MMSZ5252B
SFH615A−4
U2
4
1
R26
2.7k
R23
4.7 W
C8
0.1
100
R5
365k
R10
R11
365k
R9
R6
R7
C3
0.1uF
600V
C2
76.8k
C18
R18
R8
2k
1/2W
C4
22uF
450V
Z1
D1 − D4
1N5406 x 4
R19
MRA4007T
1.5KE440A
0.47 ”X”
1/2W
R31
3.3k
D7 MURS
160T
C6
470uF
35V
100uF
35V
C5
+
D10
Q1
1
5
6
MURS
160T
D6
R3
36k
3W
D11
R4 MMSD
100 4148T
D5
MRA4007T
R2
560k
0.5W
D8
MSR860G
100,
1/2W
11
C7
T1
100nF
8
400V 2
R24
1nF
C19 +
+
680uF,
63V x 3
+
C20 C21 C22
L3
3.3uH
C23
100,
63V
+
C24
0.1
100V
2A
48V
+
AND8394/D
Figure 3. Single Stage Converter Schematic
Single Stage Converter Characteristics
watts. CCM also reduces the high frequency output
capacitor ripple current thus improving the overall power
supply reliability. The NCP1652A controller is designed
particularly for CCM operation and also provides a second
gate drive output for the implementation of an active clamp
snubber for higher power applications where voltage spikes
caused by the flyback transformer’s leakage inductance
energy can become a significant issue. The 48 V, 2 A adapter
circuit of Figure 3 achieved an average efficiency
approaching 90% for typical operation loads. The flyback
transformer T1 was designed to effectively operate in
“deep” CCM with a relatively low magnetizing current
component (see Figure 8 for the transformer design details).
The single stage, isolated PFC converter can be derived
from a conventional buck−boost flyback topology. The
operational mode can be in discontinuous conduction mode
(DCM), critical conduction mode (CRM), or continuous
mode (CCM). While the most common operational mode
for lower power circuits is CRM, CCM offers significant
advantages for applications that require fixed frequency
operation with output voltages of 15 V dc or higher where
the use of synchronous rectification yields marginal
efficiency improvements. In CCM, the peak MOSFET
current can be significantly less than in CRM resulting in
lower switching losses, particularly at power levels above 50
http://onsemi.com
2
AND8394/D
compensation for CCM operation and can be adjusted
depending on the primary inductance and level of the CCM
transformer magnetizing ramp. The switching frequency is
set to approximately 70 kHz with C9 and should be adequate
for most applications. R5, R16, R17, D9, Z2 and C14 form
a simple OVP circuit (optional) that monitors the VCC
derived from T1’s auxiliary winding. Since this voltage will
track the output voltage, it provides a simple primary side
means of OVP sensing. R2, D5 and C4 form a DC startup
circuit for the NCP1652A which uses a peak detector to
sample the input haversine. This circuit, along with optional
TVS Z1, also doubles as a transient suppression circuit in the
event of short duration, high amplitude line transients that
input capacitor C3 would be unable to suppress.
Input EMI filter design: The EMI filter consists of two
stages of off−the−shelf common mode inductors (L1 and
L2) which are manufactured by Coilcraft. The common
mode inductors have a high leakage inductance so a single
inductor can be used for both common mode and differential
mode filtering. The differential LC low pass filter is formed
with the leakage inductance and the X capacitors from
line−to−line. The first stage of filtering is with L1 and C1,
and the second stage is with L2 and C2. L1 has a leakage
inductance of 15 mH and L2 has a leakage inductance of
22 mH.
This resulted in significantly reduced switching losses in
MOSFET Q1 over a similar CRM based design.
The single stage PFC conversion process, regardless of
the operational mode, has several compromises over the
traditional 2−stage conversion scheme of Figure 1. They are
as follows:
• As with any power factor corrector circuit, the gain
bandwidth of the control loop is very low, typically in
the 10 Hz to 30 Hz range. This is necessary, otherwise
the control loop would attempt to regulate off the
100/120 Hz line variations of the input and this would
result in a very poor power factor. As a consequence of
the low bandwidth, transient response to load step
changes will be poor although dc regulation will be
excellent. For most applications where point−of−load
regulation (POL) is utilized and/or the load is constant
(e.g. LED load), the slow transient response is
inconsequential. Figure 7 shows the output voltage
profile at supply turn−on with no load and with full
load indicating a controlled voltage rise with no
overshoot that is sometimes typical with slow control
loops.
• Because the loop cannot regulate away the 100/120 Hz
line ripple, it will appear as a ripple component on the
output. For this circuit, the output ripple was in the
order of $3% of the output voltage for full load using
with three 680 mF output capacitors (see Figure 4).
Additional output capacitance will reduce this further.
Again, for most analog or POL applications, this
magnitude of ripple should not be a problem.
• Due to the lack of a large input bulk capacitor (C3),
which would preclude high power factor, the converter
has no significant inherent hold−up time other than that
provided by the stored energy in T1 and the output
capacitors. Long hold up time is typically not a
requirement for an LED power supply.
• The power factor for the single stage converter tends to
degrade with increasing line and decreasing load due to
factors related to the D/(1−D) transfer function and
CCM operation, however, for most typical line and load
conditions the PF will be above 0.95.
Despite these tradeoffs, the single stage, isolated PFC
converter is an efficient and very cost effective solution to
many applications where the load is relatively constant
and/or point of load (POL) regulators are used downstream.
f c1 +
f c2 +
1
2p ǸL3C18
1
2p ǸL2C17
+
+
(eq. 1)
1
2p Ǹ15 mH @ 0.47 mF
+ 59.97 kHz
(eq. 2)
1
2p Ǹ22 mH @ 0.47 mF
+ 49.52 kHz
When testing in accordance with the IEC 61000−4−6
limits (0.15 MHz to 80 MHz) there should be at least −24 dB
of attenuation of the 70 kHz switching frequency with these
EMI filter component values.
On the ac input side of the filter is a fuse, F1 for safety. The
fuse is rated for 2.5 amps of continuous current, where the
average input current for 90 watts output is:
I avg +
P out Ǹ2
hV inLL
+
90 W Ǹ2
0.88 @ 90 Vac
+ 1.61 A (eq. 3)
Where h is the estimated efficiency (0.88)
The input line voltage is full wave rectified and there is a
small bulk film capacitor (C3) across the output of the bridge
rectifiers (D1, D2, D3, and D4). The capacitor is there to
decouple the high frequency switching and provide a low
impedance source for the flyback converter. A typical range
for the capacitor is 0.1 mF to 1.0 mF. For this application, a
0.1 mF capacitor was selected to minimize residual power
factor effects at light load.
T1 Flyback Transformer Design: The transformer design
for a single stage PFC converter is more complex than a
standard flyback converter and requires an iterative process,
especially when operation is in continuous conduction
mode. There are several ways to mathematically approach
this, however, treating the transformer as an energy storage
Circuit Technical Information
NCP1652A components selection: The logic level circuit
components immediately associated with the NCP1652A
shown in the Figure 3 schematic should work well with just
about any design implementation. Changes for different
output voltages involve modifying the voltage sense divider
for the TL431 (R29, R30. Zener Z4 and resistor R31 will be
required for output voltages over 32 V dc. The value of R20
will determine the peak limit of the MOSFET current and the
sub sequential maximum output current. R12 sets the ramp
http://onsemi.com
3
AND8394/D
The minimum primary inductance required is given by the
following:
Lmin = Vdc pk (min) x ton( max) / Ipk = (120 V dc x 10 ms)/
3.7 A = 324 mH
Where Vdc pk min is the low line value of the peak input
voltage (85 Vac x 1.414). Now we make an assumption: In
order to be in CCM for most of the typical loading of the
supply, let’s double this inductance value to 650 mH. The
minimum number of primary turns required for this core can
now be calculated:
choke initially is probably the most straight forward way.
Achieving the correct design on the first try is largely a
matter of experience with realizing what core volumes and
structures will support the desired power level and
accommodate the necessary coil turns. A PQ3230 ferrite
core was chosen for this design based on previous
experience and the following facts:
• The PQ core has a fairly large cross sectional area
parameter (Ae) for its overall volume. This will
minimize primary turns which contribute to unwanted
leakage inductance. The shape of the core also has a
good shielding effect on radiated emissions, particularly
if the required core gap is entirely in the center pole.
• The core window area has a good length to width
aspect ratio which helps minimize turn layers and thus
reduce leakage inductance and magnetic flux proximity
effects.
• For minimal leakage inductance, this core will
accommodate a “sandwiched” secondary winding
configuration where the secondary is in between two
primary windings which are connected in series.
Since the peak primary current in the inductor will
determine the output power, the peak magnitude of the
current can be obtained from the following relationship:
Ipk = 2 Pout / n x f x Vin (min) x ton( max)
Where h is the estimated efficiency, f is the switching
frequency, Vin(min) is the minimum average input voltage at
lowest line voltage, and ton(max) is the maximum on time.
This results in a value of:
Ipk = (2 x 95 W)/(0.88 x 70,000 x 83 V x 10 ms) = 3.7 A (pk)
Note that the average line voltage was used and not the
rms value (90 Vac) for low line. This results from the fact
that the input is a 100/120 Hz haversine and the energy
storage will be a function of the average voltage and not rms.
Now, for wire sizing purposes we need to know the rms
value of this peak primary current. If we assume an average
duty cycle of D = 0.5 V at 120 V ac input and assume an
almost rectangular CCM waveform (the magnetizing
component will actually be about 30% of the peak, but this
should give a worst case estimate), the rms value of the
switching frequency component for a pure dc input would
be:
I rms + I pk
ǸD + 3.7
0.707 + 2.6 A
Np +
L
I pk
Ae
10
B max
8
+
650 mH
1.6 cm 2
4A
10 8
2800 g
(eq. 5)
+5
Where the peak current was rounded off to 4 amps and the
maximum flux density (Bmax) was chosen to be 2.8
kilogauss to give a saturation safety margin for over−current
conditions.
Noting that the PQ3230 bobbin winding width for the
PQ3230 core is about 0.73 inches, we can comfortably get
about 30 turns of number 24 wire (dia. = 0.022”) on one layer
of the bobbin and a total of 60 turns for the entire series
primary, so 60 turns total is a good number to go with.
The next step will be to decide the primary to secondary
turns ratio. This will determine the reflected flyback voltage
on MOSFET Q1 and also the peak reverse voltage seen by
the output rectifier D8. Let’s try 2.5:1 where the secondary
will have 25 turns (120 V ac divided by 48 V dc.) With this
selection we can comfortably get 24 or 25 turns of # 24
wound over 1 layer with margins to handle the secondary
current which will be about 4 A rms worst case. The
reflected primary flyback voltage and maximum PRV
output diode voltage will occur at high line. Let’s assume an
input voltage of 270 Vac which translates to peak voltage of
378 Vdc. The reflected flyback voltage will be the peak
secondary voltage of 48 Vdc plus the output rectifier
forward drop times the turns ratio:
Vflyback = (48 V + 1 V) x 2.5 = 123 V
Adding this to the peak primary voltage of 378 V dc gives
378 + 123 = 501 V dc plus any leakage inductance spike
which will probably be in the order of 100 V. The selected
800 V MOSFET should be adequate for this. The primary of
T1 is bypassed with the voltage clamping snubber network
of D7, C7 and R3 to clamp residual spikes caused by T1’s
leakage inductance. The output rectifier’s PRV voltage will
be the peak input voltage divided by the turns ratio plus the
output voltage and output diode drop:
Diode PRV = (378 / 2.5) + 49 V = 200 Vpk (plus probable
diode recovery spike)
So, the 600 V rated ultra−fast device is more than
sufficient to handle this. Notice that a small R/C snubber
composed of R24 and C19 is across D9 to attenuate any
parasitic voltage spikes for EMI reduction. The final
transformer design is detailed in Figure 8.
(eq. 4)
Now we must divide this by 0.707 again to account for the
fact that the input to the converter is a 100/120 Hz haversine
envelope and not pure dc. This results in an rms primary
current of about 1.8 A. Looking at wire tables we can
effectively choose #24 magnet wire for the primary since it
will handle this current and should have minimal skin effect
loss at 70 kHz.
http://onsemi.com
4
AND8394/D
Test Results
mains voltages. The efficiencies were averaged and the
results are shown in the table below.
Power factor and efficiency measurements were taken at
loads of 25%, 50%, 75% and 100% at both US and Euro
Vin = 120 Vac
Vin = 230 Vac
Load
100%
75%
50%
25%
PF
0.98
0.97
0.96
0.95
h
87.0
87.2
87.1
86.1
PF
0.98
0.96
0.95
0.92
h
88.6
88.2
87.1
85
Output Ripple: The 120 Hz output ripple with a 2 A load
and 2000 mF (3 x 680 mF) of output capacitance is shown in
Figure 4 for 120 V ac input. The ripple amplitude is strictly
a function of the output capacity and load on the power
Overall Efficiency
87%
87.2%
supply since the regulation loop bandwidth is necessarily
less than the ripple frequency to assure high power factor.
Doubling the amount of output capacitance will halve the
ripple amplitude.
Figure 4. Output Ripple
http://onsemi.com
5
AND8394/D
(a)
(b)
Figure 5. Q1 Drain Waveforms at Full Load at 120 Vac and 230 Vac
Flyback Waveforms: The waveforms of Figures 5a
and 5b shows the drain voltage profile on MOSFET Q1 with
full output load (2 A) for 120 Vac and 230 Vac inputs. The
leading edge voltage spike is caused by the leakage
inductance of transformer T1 and is largely damped by the
snubber network of D7, C7 and R3.
http://onsemi.com
6
AND8394/D
(a)
(b)
Figure 6. Q1 Drain Waveforms at 25% Load at 120 Vac and 230 Vac
The waveforms of Figures 6a and 6b display Q1’s drain
waveform at 25% output load (1.25 A). Note that operation
is clearly in DCM at this load for both line input levels. Note
that when the flyback energy is depleted, the drain voltage
will ring prior to Q1’s next turn−on due to the resonant
circuit formed by the transformer’s primary inductance and
the MOSFET’s parasitic capacitance. This is a characteristic
of DCM operation.
The drain waveforms in the prototype were typical of a
relatively low leakage inductance transformer. Use of
proper PC board layout techniques with minimal power loop
traces, liberal ground planes, and clad “pours” where
possible to lower trace inductance will assure proper
switching waveform profiles and lowest EMI generation.
Very low flyback transformer leakage inductance is also
extremely important as previously mentioned. In higher
power applications and in cases were mechanical constraints
limit optimum core structure and/or winding techniques, the
use of an active clamp to suppress the primary voltage spikes
may be necessary. The optional auxiliary gate drive output
of pin 14 on the NCP1652A is provided for this function.
http://onsemi.com
7
AND8394/D
(a)
(b)
Figure 7. Output Turn−on Profiles 25% Load and at Full Load
The output turn on profiles of Figures 7a and 7b clearly
indicate that no output overshoot exists for either load
condition, particularly at light load where it is not
uncommon to exhibit overshoot with narrow bandwidth
control loops. At full load the 120 Hz throughput ripple is
clearly visible. The selection of the component values for
R28 and C26 will have a significant impact on the turn on
profile and response to load changes.
http://onsemi.com
8
AND8394/D
BILL OF MATERIALS FOR 48 V, 90 W NCP1652A ADAPTER DEMONSTRATION BOARD
Footprint
Manufacturer
Manufacturer
Part Number
Substitution
Allowed
Pb−Free
Diode
SMA
ON Semiconductor
MRA4007T
No
Yes
4
Diode
Axial Lead
ON Semiconductor
1N5406
No
Yes
D6, D7
2
Ultrafast
Diode
SMB
ON Semiconductor
MURS160
No
Yes
D9, D11
2
Signal Diode
SOD123
ON Semiconductor
MMSD4148A
No
Yes
D8
1
UFR ”soft”
Diode
TO−220AB
ON Semiconductor
MSR860G
No
Yes
Z1
1
TVS
Input Transient
Option
1.5KE440A
Yes
Yes
Designator
Qty
Description
D5, D10
2
D1, D2, D3,
D4
Value
Tolerance
Axial Lead
Z3
1
Zener Diode
18 V
5%
SOD123
ON Semiconductor
MMSZ5248B
No
Yes
Z2, Z4
2
Zener Diode
24 V
5%
SOD123
ON Semiconductor
MMSZ5252B
No
Yes
Q1
1
MOSFET
11 A, 800 V
TO−220
Infineon
SPP11N80C3
No
Yes
U1
1
PFC
Controller
SOIC16
ON Semiconductor
NCP1652A
No
Yes
U2
1
Optocoupler
4 Pin SMD
Vishay
H11A817 or
SFH6156A−4
Yes
Yes
U3
1
Progam.
Zener
2.5 V
1%
SOIC−8
ON Semiconductor
TL431A
No
Yes
C1, C2
2
X Caps
0.22 mF,
305 Vac
10%
LS = 15 mm
Rifa, Wima,
Vishay,Epcos
Box Cap − X2
Yes
C27
1
Y2 Cap
2.2 nF, 1 kV
10%
LS = 10 mm
Rifa, Wima, Vishay
Box or Disc Y2
Cap
Yes
C3
1
Polyprop.
Film
0.1 to 0.22 mF,
600 V
10%
LS =24 mm
Rifa, Wima, Vishay
Polypropylene
Film Cap
Yes (Must be
Polyprop)
C7
1
Disc Cap
68 to 100 nF,
400 V
10%
LS = 10 mm
Vishay
Ceramic Disc
Cap, 400 V Min
High Quality
Ceramic
C8, 15, 16,
17, 25
5
Ceramic
Cap
0.1 mF, 50 V
10%
1206
Anybody
100 nF, 50 V
Ceramic SMD
Cap
Yes
C24, C25
2
Ceramic
Cap
0.1 mF, 100 V
10%
1206/1210
Anybody
100 nF, 100 V
Ceramic SMD
Cap
Yes
C26
1
Ceramic
Cap
1.0 mF, 16 V
10%
1206
Anybody
1 mF, 16 V
Ceramic SMD
Cap
Yes
C19
1
Ceramic
Disc Cap
1 nF, 1 kV
10%
LS = 8 mm
Anybody
1 nF pF Ceramic
Disc Cap
Yes
C12
1
Ceramic
Cap
470 pF, 50 V
10%
1206
Anybody
Ceramic SMD
Cap
Yes
C9
1
Ceramic
Cap
680 pF, 50 V
10%
1206
Anybody
Ceramic SMD
Cap
Yes
C10, C18
2
Ceramic
Cap
1 nF, 100 V
10%
1206
Anybody
Ceramic SMD
Cap
Yes
C14
1
Ceramic
Cap
10 nF, 50 V
10%
1206
Anybody
Ceramic SMD
Cap
Yes
C13
1
Ceramic
Cap
33 nF, 50 V
10%
1206
Anybody
Ceramic SMD
Cap
Yes
C5
1
Electrolytic
Cap
100 mF, 35 V
10%
LS = 2.5 mm
Rubycon, UCC, Nich.
Radial Lead, Dia
= 5 mm
Yes
C11
1
Electrolytic
Cap
4.7 mF, 25 V
10%
LS = 2.5 mm
Rubycon, UCC, Nich.
Radial Lead,
D= 5 mm
Yes
C6
1
Electrolytic
Cap
470 mF, 35 V
10%
LS = 5mm
Rubycon, UCC, Nich.
Radial Lead
Yes
C23
1
Electrolytic
Cap
100 mF, 63 V
10%
LS = 5 mm
Rubycon, UCC, Nich.
Radial Lead
Electrolytic
Yes
C20, 21, 22
3
Electrolytic
Cap
1000 mF, 63 V
10%
LS = 8 mm
Rubycon, UCC, Nich.
Radial Lead,
D = 16 − 18 mm
Recommended
Value
C4
1
Electrolytic
Cap
22 mF, 450 V
10%
LS = 5 mm
Rubycon, UCC, Nich.
Radial Lead,
D = 12 − 15 mm
Yes
R4
1
0.25 W
Resistor
100 W
10%
Axial Lead
Anybody
1/4 W Metal Film
Resistor
Yes
R1
1
0.5 W
Resistor
1M, 0.5 W
10%
Axial Lead
Anybody
1/2 W Metal Film
Resistor
Yes
R31
1
0.5 W
Resistor
3.3k, 0.5 W
10%
Axial Lead
Anybody
1/2 W Metal Film
Resistor
Yes
http://onsemi.com
9
AND8394/D
BILL OF MATERIALS FOR 48 V, 90 W NCP1652A ADAPTER DEMONSTRATION BOARD
Manufacturer
Part Number
Substitution
Allowed
Anybody
1/2 W Metal Film
Resistor
Yes
Axial Lead
Anybody
1/2 W Metal Film
Resistor
Yes
10%
Axial Lead
Anybody
1/2 W Metal Film
Resistor
Yes
0.1 W
5%
LS = 18 mm
Anybody
Axial Lead,
Non−Inductive
Yes
3 W or 5 W
Resistor
36k to 39k
10%
LS = 30 mm
Ohmite
Axial Lead Metal
Film or WW
Resistor
Yes
1
0.25 W
Resistor
4.7 W
5%
1206
Anybody
1206 SMD
Resistor
Yes
R5
1
0.25 W
Resistor
100 W
5%
1206
Anybody
1206 SMD
Resistor
Yes
R21, R25
2
0.25 W
Resistor
10 W
5%
1206
Anybody
1206 SMD
Resistor
Yes
R27
1
0.25 W
Resistor
1k
5%
1206
Anybody
1206 SMD
Resistor
Yes
R15
1
0.25 W
Resistor
2.2k
5%
1206
Anybody
1206 SMD
Resistor
Yes
R26
1
0.25 W
Resistor
2.7k
5%
1206
Anybody
1206 SMD
Resistor
Yes
R28
1
0.25 W
Resistor
24k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R30
1
0.25 W
Resistor
5.6k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R29
1
0.25 W
Resistor
102k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R22
1
0.25 W
Resistor
10k
5%
1206
Anybody
1206 SMD
Resistor
Yes
R13
1
0.25 W
Resistor
7.32k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R14
1
0.25 W
Resistor
8.6k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R9, R12
2
0.25 W
Resistor
30.1k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R17
1
0.25 W
Resistor
39k
5%
1206
Anybody
1206 SMD
Resistor
Yes
R18
1
0.25 W
Resistor
49.9k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R19
1
0.25 W
Resistor
76.8k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R16
1
0.25 W
Resistor
100k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R10
1
0.25 W
Resistor
332k
1%
1206
Anybody
1206 SMD
Resistor
Yes
R6, 7, 11
3
0.25 W
Resistor
365k
1%
1206
Anybody
1206 SMD
Resistor
Yes
F1
1
Fuse
2.5 A, 250 Vac
TR−5
Anybody
Minifuse − TR−5
Package
Yes
L1
1
EMI Inductor
E/E Core
Coilcraft
BU10−1012R2B
No
L2
1
EMI Inductor
Toroid
Coilcraft
P3221−AL
No
L3
1
3.3 mH
Choke
3.3 mH, 5 A
LS = 0.2”
Coilcraft
RFB0807−3R3L
Yes
T1
1
Flyback xfmr
48 V, 90 W
CCM
Custom
Mesa Power Systems
13−1407 (600 mH
Primary)
No
J1, J2
2
I/O
Connectors
LS = 5 mm
DigiKey
# 281−1435−ND
(LS = 0.2”)
(for Q1,
D8)
2
Heatsink
Q1, D8
LS =
25.4 mm
Aavid
531102B02500G
(or Similar)
Designator
Qty
Description
Value
Tolerance
Footprint
Manufacturer
R8
1
0.5 W
Resistor
2k, 0.5 W
10%
Axial Lead
R2
1
0.5 W
Resistor
560k
10%
R24
1
0.5 W
Resistor
100 W
R20
1
0.5 W
Resistor
R3
1
R23
http://onsemi.com
10
Pb−Free
AND8394/D
MAGNETICS DESIGN DATA SHEET
Project: NCP1652A, 90 W, 48 Vout, isolated, single stage PFC
Part Description: CCM Flyback transformer, 70 kHz, 48 Vout
Schematic ID: T1
Core Type: PQ3230, 3C94 (Ferroxcube) or P material (Mag Inc.)
Core Gap: Gap core for 600 to 650 uH across pins 1 to 2.
Inductance: 625 uH nominal measured across primary (pins 1 to 2)
Bobbin Type: 12 pin pc mount (Mag Inc PC−B3230−12 or equivalent)
Windings (in order):
Winding # / type
Turns / Material / Gauge / Insulation Data
Primary A: (1 − 3)
30 turns of #24HN over one layer (no margins). Self−leads
to pins. Insulate for 3 kV to next winding.
48V Secondary (8 − 11)
25 turns of #24HN close wound over one layer and centered
with 2 mm end margins. Insulate with tape for 3 kV to next
winding.
Primary B: (3 − 2)
Same as primary A. Insulate for 1.5 kV to Vcc/Aux.
Vcc/Aux (5 − 6)
10 turns of #24HN spiral wound and centered with 8 mm
end margins. Insulate with tape and terminate
self−leads to pins.
Hipot: 3 kV from primary/Vcc to 48V secondary winding.
Lead Breakout / Pinout
Schematic
(bottom view)
2
7 8 9 10 11 12
30T
Pri B
3
8
25T
Pri A
30T
1
Vcc
5
6
48V sec
11
6 5 4
10T
Figure 8. Flyback Transformer Design
http://onsemi.com
11
3 2 1
AND8394/D
Conclusions
and ringing on the sync MOSFET and main MOSFET
during this transition. This usually necessitates the addition
of larger snubbers and/or TVS clamping circuits to avoid
MOSFET failure. The added circuit cost and dissipative
issues are generally not worth it.
As illustrated in this application note, the NCP1652A
when optimized for CCM operation can achieve high
efficiency across wide load variation. The use of a single
stage topology significantly reduces the overall component
bill of material cost while still achieving high power factor
in a compact form factor and well behaved startup
characteristics. Multiple protection schemes have been
implemented in this design to address ensure robust and
reliable operation.
For single stage, isolated PFC converters where the output
voltage is 15 V and less, the efficiency can be enhanced with
the use of synchronous output rectifiers instead of
convention PN or Schottky diodes. Synchronous output
rectifiers are not, however, wholly compatible with
continuous conduction mode (CCM) operation. This is
because CCM or DCM operation will almost always
transition into the other depending on the load situation. At
light load a CCM oriented design will transition to DCM and
with DCM operation, the startup and over−current mode
usually operate in CCM. As a result of these two modes of
operation, the required gate drive signal to the synchronous
MOSFET must be based on different sensing criteria for
each mode which causes additional circuit complexity. The
“problem mode” is CCM because there has to be a delayed
timing sequence to the sync MOSFET to prevent
simultaneous conduction overlap with the main primary
side MOSFET. Even though the necessary timing sequence
can be achieved, one critical issue still remains. When the
sync MOSFET is turned off just prior to the main primary
MOSFET coming on, the intrinsic body diode of the sync
MOSFET must carry the still flowing continuous flyback
current. This parasitic body diode has very poor recovery
characteristics and when the main MOSFET turns on, this
body diode is force commutated off and significant reverse
current will flow in the body diode during the recovery
process. This current along with the associated circuit
reactive parasitic components generates large voltage spikes
References
• Data sheet NCP1652
• Application note AND8124: 90 W, Universal Input,
Single Stage, PFC Converter
• Application note AND8147: An Innovative Approach
•
•
to Achieving Single Stage PFC and Step−Down
Conversion for Distributive Systems
Application note AND8209: 90 W, Single Stage,
Notebook Adaptor
Reference design TND317: 90 W Notebook AC−DC
Adapter GreenPoint® Reference Design
All the above documentation is available for download from
ON Semiconductor’s website (www.onsemi.com).
GreenPoint is a registered trademark of Semiconductor Components Industries, LLC (SCILLC).
ON Semiconductor and
are registered trademarks of Semiconductor Components Industries, LLC (SCILLC). SCILLC reserves the right to make changes without further notice
to any products herein. SCILLC makes no warranty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does SCILLC assume any liability
arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation special, consequential or incidental damages.
“Typical” parameters which may be provided in SCILLC data sheets and/or specifications can and do vary in different applications and actual performance may vary over time. All
operating parameters, including “Typicals” must be validated for each customer application by customer’s technical experts. SCILLC does not convey any license under its patent rights
nor the rights of others. SCILLC products are not designed, intended, or authorized for use as components in systems intended for surgical implant into the body, or other applications
intended to support or sustain life, or for any other application in which the failure of the SCILLC product could create a situation where personal injury or death may occur. Should
Buyer purchase or use SCILLC products for any such unintended or unauthorized application, Buyer shall indemnify and hold SCILLC and its officers, employees, subsidiaries, affiliates,
and distributors harmless against all claims, costs, damages, and expenses, and reasonable attorney fees arising out of, directly or indirectly, any claim of personal injury or death
associated with such unintended or unauthorized use, even if such claim alleges that SCILLC was negligent regarding the design or manufacture of the part. SCILLC is an Equal
Opportunity/Affirmative Action Employer. This literature is subject to all applicable copyright laws and is not for resale in any manner.
PUBLICATION ORDERING INFORMATION
LITERATURE FULFILLMENT:
Literature Distribution Center for ON Semiconductor
P.O. Box 5163, Denver, Colorado 80217 USA
Phone: 303−675−2175 or 800−344−3860 Toll Free USA/Canada
Fax: 303−675−2176 or 800−344−3867 Toll Free USA/Canada
Email: [email protected]
N. American Technical Support: 800−282−9855 Toll Free
USA/Canada
Europe, Middle East and Africa Technical Support:
Phone: 421 33 790 2910
Japan Customer Focus Center
Phone: 81−3−5773−3850
http://onsemi.com
12
ON Semiconductor Website: www.onsemi.com
Order Literature: http://www.onsemi.com/orderlit
For additional information, please contact your local
Sales Representative
AND8394/D