Journal of Physics: Conference Series

PAPER • OPEN ACCESS You may also like

- Compact atmospheric pressure plasma
Design and Implementation of Multi-output Flyback self-resonant drive circuits
V J Law and S D Anghel
Switch Power Supply - Construction of the UC2845 Simulation
Model Based on Matlab/Sumlink
Jie Zhang, Xiaokang Zhao, Zhuofan Jin et
To cite this article: Qianqiu Wang et al 2022 J. Phys.: Conf. Ser. 2396 012014 al.

- Design of a single-stage Flyback LED

Constant current Driving Power Supply
Yingchao Xu, Jiatao Lin, Xiaoqi Xie et al.
View the article online for updates and enhancements.

This content was downloaded from IP address 115.98.233.192 on 15/10/2023 at 19:45

ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

Design and Implementation of Multi-output Flyback Switch

Power Supply

Qianqiu Wang 1, Bin Yang 1, Junjie Li 1a* and Xiaoying Ye 1

1
School of Computer Science and Engineering, Neusoft Institute Guangdong, Foshan
a
[email protected]

Abstract: The switching power supply with flyback topology type has the advantages of easy
design, lightweight, and high efficiency, and it is widely used in occasions with low power and
multiple outputs. In order to solve the problems of low efficiency and low output precision of
switching power supply in the current market, this paper proposes an improved 90~264Vac to
+5V, +15V, -15V three-way output flyback with a power of 43.5W. The experimental test results
show that the proposed switching power supply has the advantages of strong anti-interference,
strong robustness, high efficiency, and low cost, and it can be widely used in various electronic
devices.

1. Introduction
The power supply is usually divided into the linear power supply and the switching power supply. A
switching power supply is a kind of power supply that integrates modern power electronics, automatic
control, power conversion, and other technologies and obtains a stable output voltage by controlling the
ratio of switch-on time. The switching power supply has rapidly replaced the traditional linear regulated
power supply with its advantages of high efficiency, small size, and wide voltage regulation range and
has become a stable power supply main force [1-3]. In the isolated topology of switching power supply,
the flyback converter has a simple structure, few peripheral components, and low cost, and is widely
used in switching power supply of medium and small power. The flyback converter converts the energy
from the primary side to the secondary side through the transformer and then outputs it to the external
load after passing through the rectifier diode on the secondary side [4-6].
In order to meet the needs of multi-voltage levels of electronic equipment, the multi-output
technology of switching power supply has always been a research hotspot [7-8]. In the multi-output
switching converter, due to the stray parameters of various components in the circuit, the output voltage
distribution does not match the turns ratio and deviates from the normal value, which brings about the
problem of multi-output cross-regulation. Aiming at solving the problems of low efficiency and low
output accuracy of the switching power supplies currently on the market, this paper proposes an
improved 90~264Vac to +5V, +15V, -15V three-way output flyback with a power of 43.5W. The flyback
switching topology power supply designed in this paper adopts UC3842 to generate frequency and
provide PWM wave, PC123S as feedback optocoupler, ER2828 as transformer model, D3SB60 as high-
efficiency rectifier tube, etc. for integrated and discrete device design. In order to further exert the
characteristics and advantages of the flyback switching topology, the output part uses multiple secondary
coils for multiple outputs and adds a first-level LDO for coordinating voltage regulation. It mainly
supplies power for the control system, drive system, various sampling circuits, etc., and has good
engineering application prospects.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

2. System structure
The overall design framework of the multi-output flyback switching power supply is shown in Figure 1.
It is mainly composed of a filter and surge suppressor, rectifier filter circuit, isolation transformer drive
circuit, the core module of UC3842, and output rectifier filter circuit. UC3842 is the core of the
switching power supply as a current-controlled pulse width modulation chip with excellent performance.
It can generate a control voltage with a fixed frequency and an adjustable duty ratio and adjust the output
voltage level by changing the on-off state of the switching power tube to achieve the purpose of voltage
regulation. The EMI filter circuit is part of the filter and surge suppressor circuit, and it is the first
interference processing circuit after the power input. Rectifier filter circuit references typical designs,
including fuse, varistor, NTC, common mode inductor, etc. The transformer is the core of the flyback
switching power supply, and its primary inductance, magnetic core air gap, winding turns, and other
parameters determine the function realization of the entire power supply. In addition, the auxiliary circuit
has sampling feedback and an isolation optocoupler.

AC DC
Filters and Surge Rectifier Isolation Secondary
Suppressors Filter Transformer Rectifier Filter

Sample
Feedback

Start-up and
Main Control Optocoupler
Power Supply Power Switch
Chip Isolation
Circuit

Figure 1. System frame diagram.

3. Working principle

3.1. Input rectifier filter part circuit

Figure 2. Schematic diagram of the input rectifier filter part.

As shown in Figure 2, CN1 is a two-port 2A connector, 3.96MM; F1 is a fuse; RT1 is a thermistor;
MOV1 is a varistor; L1 is a common mode inductor; C1 and C4 are safety capacitors (X2 capacitors);
C2 and C3 are safety capacitors (Y2 capacitors); BD1 is a rectifier bridge stack D3SB60 (3A600V); C5
is a large electrolytic capacitor.
a. Selection of fuse F1

2
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

The selection rules are as follows.

Po
F1 = 2 Irms = 2( )
γ × Vin min× PF (1)
Where γ is the power efficiency, the initial value is set to 0.8, PF is the power factor (without power
factor correction, PF=0.5; with power factor correction, PF=0.95), Vinmin is the minimum input
voltage, and the value 2 is the empirical coefficient. The value range of the fuse is 1.5 to 3 times the
theoretical value.
45
F1 = 2 × = 2.647 A
0.8 × 85 × 0.5 (2)
According to formula 2, a factor of 1.5 is selected. We finally determine that the F1 parameter is 4A,
and the fuse withstand voltage is selected as 250V (AC).
b. EMI circuit
The selection process of the common mode inductor is as follows.
• EMI test frequency: conduction 150KHZ~30MHZ, EMC test frequency: 30M~3GHZ, EMI level:
Fcc Class B.
• Determine the inductance.
1
L= = 4.61mH (3)
(2 × Pi × f ) 2 × C
Here, C2 and C3 are each 2200PF, and the cutoff frequency is initially assumed to be 50KHZ. The
actual value of L is 10mH.
• Choose the appropriate magnetic core and number of winding turns. In this design, the common
mode inductor adopts a UF10.5 skeleton, PC40 material ferrite core, and the AL value of the
high-conductivity mirror surface is 4000nH/N².
• Select the wire diameter. In this design, a wire with a diameter of 0.45 is selected.
• Choose the number of turns.
L = N 2 × AL (4)
According to the values of L and AL obtained in the previous step, N = 50T is calculated.
c. Input electrolytic capacitor calculation
1.25Po * (1 − Dch)
ΔV = (5)
Vin min* 2 * f * C

Vin min = 2 * Vin min (ac ) − ΔV (6)

Vinmax = 2 *Vin max (ac ) (7)

Dch is the capacitor charging time, generally taken as 0.2. △V usually takes 10%-15% of the rectified
voltage. We take △V=40V. It is calculated that C=93.75uF. The actual value is 100uF. After iteration,
the ripple voltage is 37.5V. Finally, we get Vinmin = 100Vdc, Vinmax = 375Vdc. Therefore, a
100uF/400Vdc electrolytic capacitor is finally selected.

3
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

3.2. Primary main circuit and clamp circuit

Figure 3. Primary main circuit and clamp circuit

a. The peak current of the MOS tube and RMS current
peak ΔI
I DS = I EDC +
2 (8)

rms peak D max

I DS = I DS ×
3 (9)
min
Po VDC × D max
I EDC = ΔI =
Where η × V × D max
min
IN, Lm × f s , D max = 0.45, Lm is the primary inductance of the
transformer, which is 1000uH.
In CMM mode,
peak ΔI
I DS = I EDC + = 1.04 + 0.45 = 1.49 A
2 (10)
In DCM mode,
peak ΔI 2 × Pout
I DS = I EDC + = 2 I EDC = = 2.08 A
2 η ×VINmin × Dmax (11)
peak peak
MOS tube usually takes 3*
I
~5* DS I
= 7.5~12.5A.
DS

The selection of the withstand voltage of the MOS tube needs to refer to the sum of the bus voltage,
the reflected voltage, and the leakage inductance peak. The calculation process is as follows. The
reflected voltage is 50V, and the leakage inductance peak voltage is preset at 100V. Finally, we choose
K410 (10A 600V) or Silan Micro SVF10N65 (TO-220F).
min max
VMOS = VDC + VOR + VRIP

= VINmax ×1.414 + 50V + 100V = 525Vdc (12)

b. RCD spike absorber circuit

The RCD snubber circuit reduces voltage spikes on the main switch caused by leakage inductance
and reduces EMI interference [9]. The parameters of each component are as follows: D1 is HER206;

4
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

R37 is a 33K/2W carbon film resistor; C33 is a 10nF high voltage ceramic capacitor with a withstand
voltage of 1KV.

3.3. Switching power transformer

The core model of the transformer is ER2828W, and the material is a PC40 ferrite core. Primary
inductance and winding turn: ER2828 core Ae=82.1, switching frequency is 60-65KHZ, then T = 16.7uS.
a. The primary inductance is
In CMM mode,
min
VDC ×D max ×T 85 ×1.414 × 0.45 ×16.7
LP1 = = = 606uH (13)
IP 1.49
In DCM mode,
min
VDC × Dmax × T 85 × 1.414 × 0.45 × 16.7 (14)
LP 2 = = = 434uH
Ip 2.08

The general transformer design adopts critical calculation. When the design is tested, the output
characteristic is optimal when the inductance is adjusted to 1000uH.
Lg
b. The core air gap is
0.4πL p I p2
δ= (15)
Ae ΔB 210 −8
c. Primary winding turns
Np =
[(Vin min*1.414)]* D max*T
ΔB * Ae
85 ×1.414 × 0.45 ×16.7
= = 68T (16)
0.16 × 82.1
Where Vinmin is the minimum voltage of AC input; Dmax = 0.45; △B is 0.16-0.25T.
d. Secondary 5V winding turns
(Vinmin *1.414)× D * NS =
(Vout + Vdiode + VR )× (1 − D )
NP (17)
Where Vout is the output voltage, Vdiode is the output rectifier diode voltage, VR is the secondary
winding voltage, and NS(5v) = 5.27T is calculated, which is set to 6T. In the same way, the 15V winding
turns are 17T, and the -15 winding turns are 20T because of the LM1915 three-terminal voltage regulator
chip behind it. The auxiliary power is 12V, then Naux = 14T.
e. Selection of primary winding wire diameter

Irms = Ipeak * D (18)

peak D
I RMS = I DS × = 2.0 × 0.39 = 0.78 A (19)
3

The skin depth is calculated as f = 7.65 . Finally, we concluded that the diameter of the wire is 2cm,
F
and the diameter of the enameled wire should be less than 0.62mm.
f. Selection of secondary winding wire diameter.

5
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

The formula for calculating the peak current of the secondary winding is I Peak
sec 2 × I OUT . The formula
=
DOFF
15 V
D , then I = 2 . 49 A
for calculating the secondary rms current is I RMS
sec sec
= I Peak ×
RMS (15V output current is
3
5V −15V
1.5A), I RMS = 4.67 A (5V output current is 3.0A), I RMS = 0.77 A (-15V output current is 0.5A).
From the RMS current of the primary winding and the secondary winding, according to the
temperature of 50°C, the current density of 4-6A and the temperature of 25°C, the current density of 6-
8A, the wire diameter and the number of strands can be calculated according to the wire area formula.
The winding method of the transformer adopts the sandwich winding method.

3.4. Secondary rectifier filter circuit

Figure 4. Schematic diagram of secondary rectifier filter

R25 is a dummy load to ensure the stability of the output no-load system. CE1, L2, and CE2 form a CLC
filter circuit to reduce the output ripple. Two parallel diodes, R and C, form a rectifier circuit, mainly to
reduce the current ripple from the secondary coil. When the entire system is shut down, the energy on
the inductor can be quickly released. The other is a three-stage steady-state circuit. The main function
of the capacitor connected to the input and output is to filter.
Selection of output rectifiers
 max N Sn 
VDIODE = 1.2 × VDC × + VOUT 
 Np 
  (1.2 is the margin factor) (20)
The effective current of the rectifier is
Iout 1 − Dmax (21)
I Drms = ×
1 − Dmax 3

The peak current of the rectifier is

2 × I Out
I Peak = (22)
1 − DMax
The conversion of 15V output is 74/17, VDCmax = 265*1.414 = 375Vdc, Nsn/Np = 17/74, Von =
15V, then Vdiode = 121.37V, the withstand voltage value of the rectifier cannot be less than 120V. The
current of the rectifier tube is not less than 6A.
1. 6 A
I Drms = × 0.428 = 1.25 A (23)
0.55

6
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

2 × 1. 6
I peak = = 5.82 A (24)
0.55
Finally, the rectifier diode selected at the 15V output terminal has a withstand voltage of 200V and a
current equal to 10A rectifier tube, and D5 selects MUR10200CT. The same method is used to calculate
the 5V one and the -15 one and choose D8 as SB1040 and D6 as HER153.

3.5. Drive circuit and output overvoltage protection

a. Determination of Oscillation Frequency
The fourth pin of UC3842 is the timing pin. After the 4th pin is grounded through the capacitor CT
and then connected to the 8th pin through the resistor RT, the oscillation circuit starts to operate. After
the switching frequency is determined, we can determine the capacitance value of the oscillation
capacitor CT and then determine the resistance value of the oscillation resistor RT. According to
f = 1.72
(RT × CT ) , R10=750K, C14=223, the approximate calculation oscillation frequency is 65KHZ.
b. Startup and Power Supply Circuit Design
The DC current voltage V charges C9 through the start-up resistors R1, R4, R30, R38, R50, R51, and
R53 and reaches pin 7. When the charging voltage reaches 16V, the chip starts to work normally.
The startup current of UC3842 is about 1mA, and the power supply current is about 30mA during
normal operation. When the startup current is 1mA, in order to enable the UC3842 to start normally, the
Vin − VCC
R7 ≤
startup resistors R1, R4, R30, R38, R50, R51, and R53 should meet I start . This design is a low-
voltage startup, and the startup capacitor parameter is C9 = 100uF.

4. Experimental results

Figure 5. Complete PCB diagram

We connect the 220v voltage to the soldered board and measure it with a multimeter. The results are
shown in Table 1. The data show that the circuit has the advantages of high efficiency and strong anti-
interference.
Table1. Circuit design measurement factors
Output Input Voltage Output Rated Output Voltage Output Ripple
Power(Max) Range Voltage Current Current Regulation and Noise
Range Accuracy
43.5W 90~264Vac 5Vdc 3.0A 0.4~5.0A 0.5% ≤50 mVP-p
15Vdc 1.6A 0.2~2.3A 2% ≤120 mVp-p
-15Vdc 0.3A 0~0.3A 2% ≤100 mVp-p

7
ICAMEE-2022 IOP Publishing
Journal of Physics: Conference Series 2396 (2022) 012014 doi:10.1088/1742-6596/2396/1/012014

5. Conclusion
When the system is under no load or light load, the proposed three-way output flyback switching power
supply system operates in a low-frequency mode, thereby effectively reducing energy consumption. It
has a variety of perfect protection measures (protection circuits), which can ensure the safety of the
system in the face of different types of emergencies. It is resistant to electromagnetic interference, small
in size, and light in weight, so it is suitable for a variety of portable devices and small power adapters.
The power supply has the advantages of simple structure, convenient use, low cost, etc. The test shows
that the circuit design function is executed normally, the operation is stable and reliable, the control
effectiveness is high, the anti-interference is strong, and it can be well applied in the power electronic
system.

References
[1] Brown M C 2012 Practical switching power supply design (Amsterdam: Elsevier) p12
[2] Pressman A 2009 Switching power supply design (McGraw-Hill Education)
[3] Maniktala S 2014 Switching power supply design & optimization (McGraw-Hill Education)
[4] Zhang J, Huang X and Wu X 2010 A high-efficiency flyback converter with new active clamp
technique IEEE Transactions on Power Electronics, 25(7): 1775-1785
[5] Coruh N, Urgun S and Erfidan T 2010 Design and implementation of flyback converters 2010 5th
IEEE Conf. on industrial electronics and applications IEEE: 1189-1193
[6] Yazdani M R, Farzanehfard H and Faiz J 2010 EMI analysis and evaluation of an improved ZCT
flyback converter. IEEE Transactions on Power Electronics, 26(8): 2326-2334
[7] Delavaripour H, Mirzaeian Dehkordi B and Adib E 2018 Dynamic model development and control
for multiple‐output flyback converters in DCM and CCM. Int. J f Circuit Theory and
Applications, 46(6): 1228-1248
[8] Chalermyanont K, Sangampai P and Prasertsit A 2007 High-frequency transformer designs for
improving cross regulation in multiple-output flyback converters 2007 7th Int. Conf. on Power
Electronics and Drive Systems. IEEE: 53-56
[9] Meng P, Wu X and Yang J 2010 Analysis and design considerations for EMI and losses of RCD
snubber in flyback converter 2010 25th Annual IEEE Applied Power Electronics Conf. and
Exposition (APEC). IEEE: 642-647