This paper introduces in detail the calculation and prediction technology of the loss of each component in the switching power supply (SMPS), and discusses the related technologies and characteristics to improve the efficiency of the switch regulator, so as to select the most appropriate chip to achieve the high efficiency target. This paper introduces the basic factors that affect the efficiency of switching power supply, which can be used as the criteria for new design. We will start with a general introduction, and then discuss the losses of specific switching elements.

1、 Efficiency estimation

    The energy conversion system must have energy consumption. Although 100% conversion efficiency cannot be obtained in practical application, a high-quality power supply efficiency can reach a very high level, with the efficiency approaching 95%

    The working efficiency of most power ICs can be measured under specific working conditions, and these parameters are given in the data. Maxim‘s data gives the data obtained from the actual test, and other manufacturers will also give the actual measurement results, but we can only guarantee our own data. Figure 1 shows a circuit example of SMPS buck converter. The conversion efficiency can reach 97%, which can maintain high efficiency even under light load.

    What secret can we use to achieve such high efficiency? We‘d better start from understanding the common problem of SMPS loss. Most of the losses of switching power supply come from switching devices (MOSFETs and diodes), and a small part of the losses come from inductors and capacitors. However, if very cheap inductors and capacitors (with high resistance) are used, the loss will be significantly increased.

    When selecting IC, it is necessary to consider the architecture and internal components of the controller in order to obtain efficient indicators. For example, Figure 1 uses a variety of methods to reduce loss, including synchronous rectification, MOSFET with low on resistance integrated in the chip, low quiescent current and pulse hopping control mode. We will discuss the benefits of these measures in this article.

    Figure 1. The MAX1556 buck converter integrates MOSFETs with low on resistance, and uses synchronous rectification to achieve a conversion efficiency of 95%. The efficiency curve is shown in the figure.

2、 Reduced voltage SMPS

    Loss is a problem faced by any SMPS architecture. Let‘s take the buck converter shown in Figure 2 as an example to discuss. The switching waveforms at each point are indicated in the figure for subsequent calculation.

Figure 2. The general buck SMPS circuit and related waveforms provide a good reference example for understanding the SMPS architecture.

    The main function of the buck converter is to convert a higher DC input voltage into a lower DC output voltage. In order to meet this requirement, MOSFET operates on and off at a fixed frequency (fS) under the control of pulse width modulated signal (PWM). When MOSFET is on, the input voltage charges the inductor and capacitor (L and COUT), through which energy is transferred to the load. During this period, the inductance current rises linearly, and the current loop is shown in loop 1 in Figure 2. When the MOSFET is disconnected, the input voltage is disconnected from the inductor, and the inductor and output capacitor supply power to the load. The inductance current decreases linearly, and the current flows through the diode. The current loop is shown in Loop 2 in the figure. The conduction time of MOSFET is defined as the duty cycle (D) of PWM signal. D divides each switching cycle into [D × TS] and [(1 - D) × TS] two parts, which correspond to the on time of MOSFET (loop 1) and the on time of diode (loop 2) respectively. All SMPS topologies (step-down, inverse equalization) use this method to divide the switching cycle and realize voltage conversion. For the buck conversion circuit, a larger duty cycle will transfer more energy to the load, and the average output voltage will increase. On the contrary, when the duty cycle is low, the average output voltage will also decrease. According to this relationship, the conversion formula of step-down SMPS under the following ideal conditions (regardless of the voltage drop of diode or MOSFET) can be obtained: VOUT=D × VIN IIN = D × IOUT should note that the longer any SMPS is in a certain state in a switching cycle, the greater the loss it will cause in this state. For the step-down converter, the lower the D (the lower the corresponding VOUT), the greater the loss generated by loop 2.

1. Loss of switching device MOSFET conduction loss

    Figure 3. The MOSFET current waveform of a typical buck converter is used to estimate the conduction loss of MOSFET? The following formula gives a more accurate method to estimate the loss, which uses the integral of current waveform I2 between IP and IV to replace the simple I2 term? PCOND（MOSFET） = [（IP3 - IV3）/3] × RDS（ON） × D = [（IP3 - IV3）/3] × RDS（ON） × In the VOUT/VIN formula, IP and IV respectively correspond to the peak and valley values of the current waveform, as shown in Figure 3? The MOSFET current rises linearly from IV to IP. For example, if IV is 0.25A, IP is 1.75A, RDS (ON) is 0.1 Ω, and VOUT is VIN/2 (D=0.5). The calculation result based on the average current (1A) is: PCOND (MOSFET) (using the average current)=12 × zero point one × 0.5 = 0.050W.

    More accurate calculation by waveform integration: PCOND (MOSFET) (calculated by current waveform integration)=[(1.753 - 0.253)/3] × zero point one × 0.5=0.089W or approximately 78%, higher than the result calculated from the average current? For the current waveform with small peak to average ratio, the difference between the two calculation results is very small, and the average current calculation can meet the requirements?

2. Diode conduction loss

    The conduction loss of MOSFET is proportional to RDS (ON), and the conduction loss of diode depends on the positive conduction voltage (VF) to a large extent. Diode loss is usually greater than MOSFET loss, which is proportional to forward current, VF and conduction time. Since the diode is conducting when MOSFET is disconnected, the conduction loss (PCOND (DIODE)) of the diode is approximately: PCOND (DIODE)=IDIODE (ON) × VF × In (1 - D) formula, IDIODE (ON) is the average current during diode conduction. As shown in Figure 2, the average current during diode conduction is IOUT. Therefore, for step-down converter, PCOND (DIODE) can be estimated as follows: PCOND (DIODE)=IOUT × VF × (1 - VOUT/VIN) is different from MOSFET power consumption calculation. Using the average current can obtain more accurate power consumption calculation results, because the diode loss is proportional to I, not I?. Obviously, the longer the conduction time of MOSFET or diode is, the greater the conduction loss is. For buck converters, the lower the output voltage, the greater the power consumption generated by the diode, because it is in the on state for a longer time.

3. Switching dynamic loss

    Since the switching loss is caused by the non ideal state of the switch, it is difficult to estimate the switching loss of MOSFET and diode. It takes a certain time for the device to fully turn on or turn off or turn on, and power loss will occur in this process.

    The diagram of drain source voltage (VDS) and drain source current (IDS) of MOSFET shown in Figure 4 can well explain the switching loss of MOSFET in the transition process. It can be seen from the waveform in the upper half that voltage and current transients occur during tSW (ON) and tSW (OFF), and MOSFET capacitors are charged and discharged. As shown in Figure 4, VDS drops to the final on state (=ID × Before RDS (ON), full load current (ID) flows through MOSFET On the contrary, the VDS gradually rises to the final value of the off state before the MOSFET current drops to zero during the off state. During switching, the overlapping part of voltage and current is the source of switching loss, which can be clearly seen from Figure 4.

    Figure 4. The switching loss occurs during the transition process of MOSFET on and off. The switching loss increases with the increase of SMPS frequency, which is easy to understand. With the increase of switching frequency (cycle shortening), the proportion of switching transition time increases, thus increasing the switching loss. In the switching process, when the switching time is one twentieth of the duty cycle, the effect on efficiency is far less than when the switching time is one tenth of the duty cycle. Because switching loss is closely related to frequency, switching loss will become the main loss factor when working at high frequency.