Optimization design of DC-DC power supply system

With the miniaturization of electronic devices, the power supply of compact electronic devices is a very important issue. At present, DC-DC converters are widely used in battery-powered devices and compact electronic devices that require power saving. The purpose of using DC-DC converters is to convert voltage and provide suitable operating voltage for some devices, but more importantly, to ensure high system efficiency and small size while converting voltage. Under normal circumstances, excellent DC-DC converters have a conversion efficiency of more than 95%. Higher system efficiency can not only extend the battery life, but also further reduce the size of the device.

After analysis, it is not difficult to find that the system efficiency of DC-DC power supply is limited by the energy-consuming components of the power supply system itself, such as power supply internal resistance, filter impedance, connecting wires and contact resistance, etc. On the other hand, it is also closely related to the working state of the DC-DC converter and the power supply parameters. Reasonable configuration of these design parameters can improve the system efficiency. The energy consumption of the power supply internal resistance will reduce the efficiency of the power supply itself, and also affect the input voltage of the DC-DC converter, thus affecting the conversion efficiency of the DC-DC converter. In extreme cases, the DC-DC converter will enter an abnormal state. In severe cases, the system will stop working completely. Even if it can work normally, the system efficiency will be seriously lost. Therefore, in the design, the reasonable selection of power supply voltage, the reduction of power supply internal resistance, and the correct selection of the working point of the DC-DC converter can effectively improve the system efficiency of the DC-DC power supply. The key to the optimal design of the DC-DC power supply system is to correctly analyze the interaction between the various parts of the electronic equipment (especially between the power supply and the DC-DC converter) and find out the main factors affecting the efficiency of the power supply system.

1 Power distribution in general electronic devices

Generally, electronic systems containing DC-DC power supplies can be divided into three parts: power supply, voltage regulator (DC-DC converter) and load, as shown in Figure 1. The actual power supply part can be a battery pack or a regulated or unregulated DC power supply. When analyzed, it can be equivalent to an ideal voltage source Vs and a power supply internal resistance Rs. Among them, Rs includes the power supply output impedance, series filter resistance, wire resistance and contact resistance, etc. These resistances are energy-consuming components and will seriously affect the power supply efficiency. The power supply efficiency (Es) is defined as the ratio of the power absorbed by the voltage regulator to the total power provided by the power supply: The voltage regulator consists of a control IC and related peripheral components. Some characteristic parameters of the control IC can be found in the data sheet provided by the manufacturer. The conversion efficiency (Ed) of the voltage regulator is defined as the ratio of the power provided by the DC-DC converter to the load part to its input power: According to the manufacturer's description, the conversion efficiency Ed of DC-DC is a function of the input voltage Vi, the output voltage Vo and the load current Io. However, under normal circumstances, the conversion efficiency Ed is not sensitive to changes in the load current Io. When the load current Io changes by more than two orders of magnitude, the efficiency change will not exceed a few percentage points. Therefore, under non-extreme conditions, the conversion efficiency of the DC-DC converter can usually be approximated as a constant. However, under extreme conditions, the conversion efficiency will be severely lost, which can be seen from the input characteristics of the DC-DC converter (as shown in ① in Figure 2). Here, the DC-DC converter is temporarily regarded as a two-port black box.

The load part is similar to the power supply part, and can also be equivalent to the effective load Rl and the energy-consuming element Rp connected to it. The efficiency of the load part (Ei) is defined as the ratio of the actual power absorbed by the effective load to the output power provided by the converter: the efficiency of the entire system should be the product of the efficiencies of Es, Ed, and El. Since these three parts affect each other, the key to the optimal design of the system lies in correctly analyzing the interaction between the three and reasonably configuring the relevant parameters between them to maximize the efficiency of the entire system.

2 Key factors affecting system efficiency

Assume that the system efficiency is Ea, then Ea=EsEdEl

From equations [1], [2], and [3], we can get: In the actual system, Vo and Vl are determined by the load circuit requirements, and the power supply should provide a suitable stable voltage to the load. Due to the special design inside the DC-DC converter, Ed only changes within a small range under normal conditions. As long as the operating point is selected appropriately, Ed is approximately a constant. Only Vi and Vs are optional, so the key to ensuring high system efficiency lies in the values ​​of Vi and Vs. And Vi and Vs depend on the interaction between the power supply and the voltage regulator. According to the circuit principle, there is such a relationship between them:

Vs=IiRs+Vi (5)

Obviously, in order to reduce the loss of the power supply internal resistance, after Vs is selected, Ii should be as small as possible, while Vi should be as large as possible, which coincides with the input characteristics of the DC-DC converter (as shown in ① in Figure 1). From the input characteristics of the DC-DC converter, it can be seen that within the effective working range, a larger Vi means that the regulator will draw a smaller current, so that the loss on the power supply internal resistance will be reduced, while the output remains almost unchanged. This improves the efficiency of the power supply part. So how can the regulator draw a smaller current from the power supply while the output remains almost unchanged? This requires a reasonable selection of the working point of the DC-DC converter.

3. Selection of DC-DC converter operating point

In Figure 2, ① is the input characteristic curve of a general DC-DC converter. It can be seen that its input has a certain dynamic range, and its input characteristic curve can be clearly divided into three intervals. When Vig≤Vi≤Vmin, the DC-DC converter is in a transitional state of startup operation. During this period, the current drawn from the power supply by the DC-DC converter rises rapidly with Vi until the input current reaches the maximum value Iimax when the output reaches the set value; the DC-DC converter may work in this section, but the system efficiency will be very low, including the power efficiency and the conversion efficiency of the DC-DC converter. The Vminis interval is the effective working area of ​​the DC-DC converter. In this interval, the DC-DC converter has a higher and more stable conversion efficiency, so comprehensively speaking, the working point of the regulator should be selected at the high end of this section.

In Figure 2, ② is the resistive load characteristic of the power supply. It is determined by equation (5). The intersection of ① and ② is the working point Q of the DC-DC converter. In order to make ① and ② have an intersection and fall within the effective working area of ​​the DC-DC converter, Vs and Rs must be selected reasonably. Rs determines the slope of the characteristic equation, and Vs determines the intersection of the characteristic equation and the horizontal axis. Therefore, changing Vs and Rs can move the working point Q. Combined with the above analysis, the working point should have a higher Vi as much as possible.

In addition, it can be seen from the figure that when Vs is selected, the operating point is determined by the internal resistance Rs of the power supply. In order to ensure that the operating point never enters the non-effective working area, the load line slope (-1/Rsmax) should have a limit, that is, the internal resistance Rs of the power supply should have an upper limit Rsmax. It can be seen from the figure: that is to say, the total resistance Rs between the power supply part and the DC-DC converter should be guaranteed to be always less than Rsmax. Otherwise, the operating point of the DC-DC converter will enter the abnormal working area and seriously lose the system efficiency, or even make the DC-DC converter stop working completely. This is particularly important in actual design. The DC-DC power supply has very high requirements for the resistance Rs between the power supply part and the DC-DC converter. For example, if a 5V power supply is converted into a 3.3V output and a load current of 2A is provided, if a DC-DC converter MAX797 chip (Vmin=4.5V) is selected and a 90% conversion efficiency is guaranteed, then Rs should not be greater than 0.307Ω; if a 95% conversion efficiency is required, then Rs should not be greater than 0.162Ω. It can be seen that the DC-DC power supply has very high requirements for the resistance Rs between the power supply part and the DC-DC converter. Rs is also a key factor affecting the system efficiency.

Based on the above analysis, the efficiency of DC-DC power supply system in compact electronic devices is a very important issue. The key to the optimal design of DC-DC power supply system lies in correctly analyzing the interaction between power supply and voltage regulator, and reasonably configuring the parameters of power supply and the working point of regulator, which can effectively improve the efficiency of the whole system.