Non-isolated step-down LED driver power supply
The driving method of LED is different from traditional halogen lamps and fluorescent lamps. It needs to maintain constant current driving, so special driving power is needed. As general lighting, most of them are high-voltage mains input and SELV (safe extra-low voltage) output, so they mostly use step-down structure. Buck topology has the characteristics of simple structure, high efficiency, and small current ripple. It is often used. . PT4207 is an LED driver chip designed based on Buck topology.
PT4207 chip structure characteristics
PT4207 adopts an innovative architecture, which can work reliably under the DC voltage of 8V to 450V after AC input is rectified. Built-in 350mA/20V MOSFET can provide 350mA LED output current. In addition, it is equipped with an external MOSFET switch drive port to achieve The LED output current is up to 1A and works stably. The system efficiency can reach 96%, and the LED current accuracy can reach ±5% (including input voltage adjustment rate and component differences). Through the multi-function dimming DIM pin, the LED current can be linearly adjusted using resistance or DC voltage, or the digital pulse signal can be used to select PWM dimming. In addition, the chip also has soft-start, short load and over-temperature functions. The internal structure block diagram of PT4207 is shown as in Fig. 1.
Figure 1PT4207 internal structure block diagram
Constant current working principle: PT4207 uses a fixed off time mode to control the output current. After the internal MOSFET, the current flows through the load, inductance, MOSFET and sampling resistor, and linearly rises with time, and a voltage is generated at the CS pin. When the voltage reaches the internal reference value, the chip internally controls the power to turn off the MOSFET and enters the turn-off cycle . The turn-off time is set by an external resistor and is fixed. After the lapse, the MOSFET turns on again and enters the next working cycle. The way of Buck structure is shown in Figure 2.
Figure 2 Two forms of Buck structure
During the MOSFET turn-off period, the energy in the inductor L is released into the load LED through the freewheeling diode D, and is formed back, as shown in Figure 3.
Figure 3 Buck structure turns off cycle current return
can be obtained by the inductance formula
where VL is the voltage across the inductor, L is the inductance, Toff is the settable fixed off time, and ΔIL is the amount of current in the inductor.
Figure 4 Inductor current waveform under CCM
If the system is working in CCM (continuous working mode), the current waveform in the inductor is shown in Figure 4. Among them, ILED is the LED uniform current, IPEAK is the peak current in the inductor, that is, the peak current through the MOSFET or freewheeling diode, and ILED=IPEAK-0.5ΔIL is obtained. Substitute the inductance formula to obtain
IPEAK can be set by sampling resistor. Therefore, once the output LED scheme is determined, the output current has nothing to do with the input voltage, thus realizing LED constant current control.
Short principle: The chip detects the CS pin voltage in each turn-on cycle. Once it detects that the CS voltage rises too fast, the chip will turn off the MOSFET and turn it on again after a period of time to achieve short.
Over-temperature principle: The chip has a built-in overheating function. When the junction temperature of the chip exceeds 135°C, the output current will be automatically reduced to further increase the temperature. If the temperature exceeds 150°C, the output current will drop to 0, which can avoid flickering problems while the chip is active. If you need to over-temperature the LED, you can indirectly connect a negative temperature coefficient thermistor between the DIM pin and the GND pin. When the temperature rises, the DIM voltage will drop, and at the same time reduce the internal CS pin reference voltage or even shut down, so as to achieve Over temperature function.
Soft start energy: The chip has a built-in 4ms soft start time, and the current is gradually increased when starting, so that the load current gradually reaches the set value, effectively reducing the starting surge current.
Figure 5PT4207 typical application power (output: 24 strings of LED array, 250mA) (print)
Figure 6 PT4207 typical application electrical efficiency and constant current characteristics
Figure 7PT4207 high current application (output 12 strings of LED array, 1000mA)
Figure 5 is a typical application of PT4207. The efficiency and constant current characteristics of the typical application of PT4207 are shown in Figure 6. Other application schemes of PT4207 are shown in Figure 7 and Figure 8. Among them, Figure 7 is the high current application of PT4207 (output 12 strings of LED array, 1000mA); Figure 8 is the PT4207 DC low voltage application (output 1 3WLED, 700mA).
Figure 8PT4207 DC low voltage application (output 1 3WLED, 700mA)
System parameter design
Refer to Figure 5 for typical applications. The determination of the output current: can be based on the formula
Select the appropriate R4, R5, R6 and L. For specific calculation steps, please refer to the PT4207 data sheet.
Input capacitance selection: The input capacitance provides a stable power supply voltage for the system, which can be selected according to the output power and capacitance according to 1-2uF/W. Lighting applications are all in high temperature, so the temperature resistance of the capacitor is above 105°C.
MOSFET selection: the drain-source withstand voltage Vds is selected according to the actual input situation, and the drain current Id is 4 times or more ILED.
Output capacitor selection: The capacitor connected in parallel with the LED can absorb the LED ripple current. Ideally, the inductor ripple current is completely absorbed by the output capacitor, extending the life of the LED to a certain extent. Usually choose 1-10uF.
Freewheeling diode selection: Choose Schottky diode or ultra-fast recovery diode, the reverse recovery time Trr is less than 100ns, and the current capability should be greater than IPEAK.
LED fluorescent lamp shell inductance selection: I-shaped inductor or closed magnetic transformer inductor can be selected. I-shaped inductors are generally low in price and simple in process, but they are magnetic, which can easily cause the loss of magnetic lines in a metal confined space and cause the system to work abnormally, so they are generally used in lamps with non-metal shells. No matter which kind of inductor is used, the saturation current of the inductor is required to be greater than 1.2 times the ILED, and the Curie temperature of the magnetic core material is greater than 150°C.
Layout design points
Refer to Figure 5 for typical applications. Among them, the filter capacitors C3, C4, C5, and resistor R4 should be as close as possible to the chip pins. Input capacitor C1, load, inductor L4, MOSFET, chip S pin, sampling resistors R5 and R6 are large current paths, the wiring should be as thick and short as possible, and the enclosed area should be as small as possible. Sampling resistors R5 and R6 are connected to high-frequency and high-current ground, which are interference sources and should be connected to the negative electrode of the input filter capacitor C1 through the shortest path. The third pin of the chip, as well as the ground of C3, C4, C5, and R4 need a stable reference ground, which can be led out separately from C1.
