1. Drive power topology and control method
The driving power required by the LED is rectified by the alternating current and then directly converted. The rectifier circuit is usually diode-bridge rectified and filtered by an electrolytic capacitor. This method has a relatively low power factor and brings a large harmonic pollution to the power grid. The active power factor correction circuit reduces the pollution of the harmonics to the power grid. Therefore, the topology of the power supply should be able to achieve PFC better, and the loss is also an important factor to be considered. Finally, the power supply of the LED usually needs to be closed. Size is also limited. Therefore, the selected converter should have the following advantages: less device, high efficiency, and small size. Commonly used active power factor correction topologies are BOOST, flyback converter, SEPIC, etc. The BOOST converter is simple and has high efficiency, but it can only achieve boosting. It is suitable for occasions where the output voltage is higher than the input voltage. The LED driving power supply needs to be boosted/depressed, so BOOST cannot be used. The isolated flyback converter can also achieve power factor correction. The output voltage can both boost and step down, but the transformer in the flyback converter only works in the first quadrant, the core utilization rate is not high, and it needs to be added. With some buffer circuits, the efficiency of the converter is not high, and the power supply size is large. The output of the SEPIC circuit can achieve boost and buck, and the input current of the SEPIC converter is continuous relative to the flyback converter. The input inductor used for filtering is small, and the SEPIC does not need to add a buffer circuit. By reducing the size of the power supply and increasing the efficiency of the power supply, the SEPIC circuit is selected as the topology of the drive circuit.
Figure 1 is a simplified diagram of the main circuit and control circuit of a high power LED lighting power supply based on a SEPIC converter. The brightness of the LED is roughly proportional to the amount of current flowing through the LED. The brightness of the LED can be adjusted by controlling the current of the LED. The voltage on C1 in Figure 1 is the voltage after bridge rectification. R1 and MOS are connected in series to sample the current flowing through the MOS transistor. R2 and the load LED are connected in series to sample the load current signal.
As can be seen from Figure 1, R2 compares the current flowing through the LED, and the obtained signal is compared with the reference signal Vref. The error is amplified by the amplifier and used as an input to the multiplier to control the brightness of the LED and change the sampling resistance. The size of R2 can change the brightness of the LED. The other input of the multiplier is the sampling signal of the input terminal voltage, and the multiplier output result is compared with the current sampling signal of the MOS tube and the inductor L1, and the generated PWM pulse is used to control the switch of the MOS tube to realize the load current and the input. Current control, which ultimately completes LED brightness adjustment and power factor correction.
2. Analysis of SEPIC working principle
According to whether the current flowing through D5 is always greater than zero, the working mode of the SEPIC circuit is divided into intermittent working mode, continuous working mode (CCM) and critical continuous working mode, and PFC is realized by BCM. The equivalent circuit in different switching modes in critical continuous mode is shown in Figure 2. In the following analysis, Ts represents the switching period, and Ton and Td are times when the MOS transistor is turned on and the diode is turned on in one cycle.
1) Working mode 1: MOS tube is opened
The equivalent circuit diagram of the SEPIC circuit when the MOS transistor is turned on. At t=0, the MOS transistor Q is turned on, and the diode D5 is turned off. In the figure, the voltage vc1 of C1 is used as the power supply voltage, which is a pulse wave after diode bridge rectification, and the peak value of the pulse wave is represented by VC1. Since the switching frequency is much larger than the bus frequency, the bus voltage can be considered constant during one switching cycle, that is, the voltage vc1 on C1 is considered constant, and the input is a DC signal. At this time, two loops are formed: the first one is the power supply C1, L1 and Q loops. Under the action of vc1, the inductor current iL1 linearly increases, and the current waveform is as shown in Fig. 3(a). The second is the C2, Q and L2 loops. The inductor current iL2 increases linearly, while C3 supplies power to the load. The current waveform is shown in Figure 3(b). Assume that the current flowing through the inductor L1 and the current flowing through L2 at time t=0 are iL1(0) and iL2(0), respectively. When Q is turned on, the voltage applied to L1 is vc1, which proves that when the C2 size is selected properly Vc2=vc1, the voltage on L2 is also vc1. Can get
Where 0 ≤ t ≤ Ton. It can be seen from the above three equations and FIG. 3 that when t=Ton, iL1(t) and iL2(t) are the largest, when the MOS transistor is turned off, and the working mode is over, the current waveform on the MOS transistor is as shown in the figure. 3(b).
2) Working mode 2: diode conduction
Figure 2(b) shows the equivalent circuit diagram of the SEPIC circuit when the MOS transistor is turned off. When t=Ton, the MOS transistor Q is turned off. At this time, two loops are formed. The first one is that the power supplies C1, L1, and C2 pass through the diode D5 to the load, and the power supply and the inductor L1 store energy simultaneously to the C2 and the load, and the C2 energy storage. Increase, and iL1 decreases; in addition, L2 passes D5 to the load loop, L2 storage energy is released to the load, so iL2 falls, and the current waveform is shown in Figure 3(a), (b). Since D5 is turned on, the voltage applied to L2 is -V0, where V0 is the output voltage, and the voltage on C1 is equal to the input voltage, so the voltage applied to L1 is also -V0, flowing when iL1=-iL2 The current of diode D5 drops to zero, the diode turns off, and the diode current waveform is shown in Figure 3(d). At this time, the MOS transistor Q is turned on, and the circuit operates in a critical continuous mode. According to the above analysis, the diode conduction phase can be obtained.
The control loop compensation parameters are adjusted by selecting appropriate R3, R4, C4 and C5 values ​​such that the bandwidth of the entire control loop of the power supply is less than 20 Hz and lower than the line voltage frequency, and the output of the compensator can be considered to be 1/2 The power frequency period is constant, so the peak current of the MOS tube is proportional to the line voltage, the peak current of the MOS tube is also a sinusoidal curve, and the peak value of the sinusoid is expressed by Ipk, and the peak value of the MOS tube current can be obtained.
In the critical continuous mode, according to formula (1) and formula (2)
Where ton(t) is the MOS tube conduction time in each switching cycle within a half-frequency cycle period. When the MOS transistor Q is turned on, the peak current flowing through the MOS transistor can be obtained according to the equation (3).
It can be seen from equation (12) that when the SEPIC circuit operates in the critical continuous mode, the turn-on time of the MOS transistor is fixed under certain input voltage and load conditions. According to L1, the volt-second balance on L2 can be obtained by td(t)=Tonvc1(t)/V0, and the switching frequency of the MOS tube can be obtained.
It can be seen that the switching frequency of the critically continuous SEPIC circuit varies with the input voltage, which is different from the DCM operating at a constant switching frequency.
Considering that the circuit operates in the critical continuous mode, the current flowing through the MOS transistor when the MOS transistor is just turned on is 0, which can be obtained according to the ampere-second balance on C2.
It is an ideal sine wave with a power factor of 1. At the same time, when K is small, the power factor can be made close to 1.
3, SEPIC circuit parameters and experimental results
The specific circuit parameters in the experiment are: input voltage amplitude range: AC85 ~ 265V; L1 = 1.4mH; L2 = 0.45mH; C1 = 10nF; C2 = 0.47μF; C3 = 680μF; LED is 20 high-bright white LEDs in series. Control loop compensation device parameters: R3 = 22kΩ; R4 = 27kΩ; C4 = 1μF; C5 = 33nF. Output current: 350mA. Power factor: 0.9 or more. Output voltage ripple: "5%.
Figure 4 shows the phase relationship between the input voltage and the input current. The input voltage in the figure is the grid voltage after the transformer is stepped down. It can be seen from Fig. 4 that the input current is in the same phase as the input voltage, and the input current follows the input voltage very well, realizing the power factor correction.
Figure 5 shows the input voltage of the diode bridge rectified, the driving voltage of the MOS transistor, the current waveform of the inductor L1 and the inductor L2, wherein Fig. 5(a) is a global map, Fig. 5(b) and Fig. 5 (c) is a partial enlarged view.
As can be seen from Fig. 5(a), the envelope of the current waveform on the inductor L1 and the inductor L2 is the positive half cycle of the sinusoidal alternating voltage. The experimental waveforms of Figures 5(b) and 5(c) and the theoretical analysis of Figure 3
It can be seen from the comparison between Fig. 5(b) and Fig. 5(c) that the turn-on time of the MOS transistor is constant, the turn-off time is variable, and the switching frequency is also variable, which is also controlled by the DCM. A difference.
Figure 6 shows the input voltage, output voltage ripple and output voltage after diode bridge rectification. It can be seen that when the output DC voltage is 65V, the ripple voltage peak-to-peak value is 2V, and the output ripple is about 3. %, the output ripple is small.
Figure 7 shows the power factor and efficiency curves of the SEPIC converter for different input voltages. It can be seen that under certain output voltage conditions, the input voltage becomes higher and the power factor is gradually lower. This conclusion is consistent with the theoretical analysis of the relationship between power factor and input voltage for SEPIC operation in critical continuous mode. Although the power factor is reduced when the voltage is high, they are all above 0.95, which achieves the purpose of power factor correction, and the overall power efficiency is as high as 92.3%.
4, the conclusion
This paper introduces a design of high power factor power supply for LED illumination. The main circuit topology of the power supply adopts SEPIC converter, which uses single-stage converter to realize power factor correction. It uses less devices, has low loss, and has small power supply volume. Feedback control Simple, it can boost and step down the output voltage and control the output current to achieve brightness control of the LED. Firstly, it is proved theoretically that the SEPIC converter can realize power factor correction when operating in critical continuous mode. The relationship between power factor value and input-output voltage ratio is analyzed. Then the experimental results show that the input voltage is between 85 and 265V. The values ​​are all above 0.95, which achieves the purpose of power factor correction, and can control the output current to achieve the control of LED brightness. The efficiency of the whole power supply is as high as 92.3%.
SHAOXING COLORBEE PLASTIC CO.,LTD , https://www.colorbeephoto.com