The flyback topology is the most commonly used topology for designing multi-output discrete switching power supplies and 48V input telecommunications applications. The following describes how to design a low-cost discontinuous flyback switching regulator using the high-efficiency LM3488 low-side N-channel current-mode controller. A flyback converter is essentially a buck/boost converter. A single inductor can be divided into two parallel inductors with a 1:1 turns ratio. Separating them does not change the basic function of the inductor; the two parallel coils of the same core are equivalent to one coil. If the connection between the two coils is broken, current flows through coil L1 when the transistor conducts, but current flows through coil L2 when the diode conducts. The points on the primary and secondary coils of the transformer are exactly opposite in position; this characteristic immediately confirms that a flyback topology is used. The biggest problem with this topology is the design of the inductor-transformer, as this inductor-transformer is responsible for both storing energy and performing transformer functions. Unlike a typical transformer in theory, current does not flow into both coils simultaneously. The primary coil that generates magnetization uses the same inductance as the inductor in a traditional buck/boost converter. When the transistor conducts, power from the DC power supply is stored in L1. When the diode conducts, power is transferred to the output capacitor and the load. Comparison of Discontinuous vs. Continuous Mode: Flyback converters have two different operating modes: discontinuous mode and continuous mode. Both modes have the same circuit diagram. The current waveforms of the primary and secondary coils of the transformer are shown. If the output current increases above a certain level, the circuit, originally designed for discontinuous mode, will switch to continuous mode. In discontinuous mode, all the energy stored in the primary coil during startup is transferred to the secondary coil and the load before the next cycle begins. There is also a brief period of inactivity when the secondary coil current drops to "0" before the next cycle begins. In continuous mode, some power is retained in the secondary coil at the start of each new cycle. Flyback converters can operate in either mode, but each mode has different functional characteristics. Discontinuous mode has a higher peak current, allowing for a higher output spike voltage after the switch is turned off. However, it also has a faster load transient response and a lower primary inductance, allowing for a smaller transformer size. Since the forward current before the input reverse voltage is "0", the reverse recovery time of the diode is less critical. Furthermore, the collector current of the transistor is "0" at startup, helping to reduce electromagnetic interference noise in discontinuous mode. As for continuous mode, although its peak current is lower, resulting in a lower output spike voltage, it is rarely used in low-power applications because the right half of the converter's transfer function is "0", necessitating significant bandwidth compression to stabilize the feedback loop. A comparison of voltage-mode and current-mode control: The voltage-mode control circuit uses only one feedback loop. A pulse-width modulation circuit using this mode is shown. The oscillator in the diagram generates a fixed sawtooth triangular waveform Vst using the fixed current of an external capacitor. The error amplifier compares the output voltage feedback with the reference voltage and generates an error voltage Ve based on the comparison result. The voltage comparator compares Ve with Vst. If Vst is higher than the error voltage Ve, the pulse width modulation output will also be higher. Voltage-mode control cannot control the output current; therefore, load transients must be sensed through output voltage changes and corrected using a feedback loop. The advantage of current-mode control is that it can control both the output current and output voltage of the same circuit. Line and current transient responses can be provided simply by changing the duty cycle of the power transistor. The output voltage error Ve and the sawtooth waveform Vst both determine the duty cycle, while the sawtooth waveform Vst is generated by the output inductor current from an external current-sensing resistor. Since current-mode control provides current limiting per cycle, as well as excellent bandwidth and transient response, it is a more ideal choice. Basic Operation This is a typical flyback converter design using the LM3488. Vo1 is the master output of this circuit, while Vo2 and Vo3 are slave outputs. The slave outputs adjust with line changes to ensure voltage stability and also adjust to load changes to ensure voltage stability. When Q starts, Np generates a fixed voltage, and the current rises linearly, which can be expressed by the following formula: dI/dt = (Vin 1/Lp), where Lp is the magnetizing inductance of the primary coil. When startup is complete, the primary coil current has risen to Ip-peak = (Vin 1)Ton/Lp. During startup (Ton), the power stored in the inductor can be expressed by the following formula: E:={L_{p} (I_{p-peak})}︿{2}\over{2} When Q is turned off, the inductance of the primary coil reverses the polarity of all coils. Assuming only one output, theoretically all the energy of the primary coil should be transferred to the secondary coil at the instant of shutdown. The maximum current of the secondary coil is equal to Is-peak = Ip-peak N, where N is the turns ratio between the primary and secondary coils (Np/Ns). The electrical power transferred to the output is: P:={E}\over{T}={[（V_{in}-1） T_{on}]}︿{2}\over{2 T L_{p}} where T refers to the time equal to 1/switching frequency, and Ton refers to the on-time. The flyback loop fixes the product of Vin and Ton at a certain level to ensure that the output voltage remains stable. In transformer design theory, transformers should not store energy; all energy should be immediately transferred from the primary coil to the secondary coil. A flyback transformer can be used as an energy storage device. After the switch is started, most of the energy is stored in the primary inductance of the transformer. After the switch is closed, the energy is transferred to the secondary coil, the output capacitor, and the load. Energy is stored in the air gap of the core; if a permeable alloy powder core is used, the energy is stored inside the core. Inductor transformers are designed to minimize leakage inductance, AC coil losses, and core losses. Leakage inductance is the primary inductance that does not couple with the secondary inductance. Because leakage inductance reduces transformer efficiency and generates peak values ​​at the drain of the switching chip, it must be minimized. High coil losses are primarily caused by the skin effect. The higher the frequency, the more current tends to flow towards the surface of the conductor; therefore, braided or sheet coils are generally used. Braided coils are typically made by twisting together several thin wires, and several braided coils can be further twisted together to form a thicker strand. The core window should be as wide as possible to reduce the number of layers, minimizing AC coil losses and leakage inductance. Class E cores with internal air gaps are the preferred solution for low cost and low leakage inductance. Core losses depend on the core material, switching frequency, and current swing. For flyback transformers employing discontinuous conduction design and switching frequencies exceeding 100 kHz, ferrite P material is typically the preferred choice due to its ability to reduce core losses. The LM3488 driver operates over a wide frequency range, from 100 kHz to 1 MHz. When selecting the operating frequency for a power supply, careful consideration should be given to factors such as switching losses, overall transformer losses, the size of magnetic components, cost, and output capacitors to maximize the advantages of each. Higher switching frequencies reduce output capacitance and the inductance of the primary and secondary coils, thus helping to reduce transformer size. However, higher switching frequencies also increase transformer losses and switching losses of the switch. High losses reduce the overall efficiency of the power supply, necessitating a larger heatsink to dissipate the accumulated heat.