Switching power supply PCB design principles and routing techniques

1. Introduction

Switching power supply is a voltage conversion circuit. Its main working content is to step up and down the voltage. It is widely used in modern electronic products. Because the switching transistor always works in the "on" and "off" states, it is called a switching power supply. The switching power supply is essentially an oscillating circuit. This way of converting electrical energy is not only used in power supply circuits, but also widely used in other circuits, such as the backlight circuit of liquid crystal displays and fluorescent lamps. Compared with transformers, switching power supplies have the advantages of high efficiency, good stability and small size. The disadvantages are that the power is relatively small and it will produce high-frequency interference to the circuit. The transformer feedback oscillation circuit is a circuit that can generate regular pulse current or voltage. The transformer feedback oscillation circuit is a circuit that can meet this condition.

Switching power supplies are divided into two forms: isolated and non-isolated. Here we mainly talk about the topological form of isolated switching power supplies. In the following text, unless otherwise specified, it refers to isolated power supplies. Isolated power supplies can be divided into two categories according to different structural forms: forward and flyback. Flyback means that when the primary side of the transformer is turned on, the secondary side is cut off and the transformer stores energy. When the primary side is cut off, the secondary side is turned on, and the energy is released to the load. Generally, conventional flyback power supplies are mostly single-tube, and double-tube ones are not common. Forward means that when the primary side of the transformer is turned on, the secondary side senses the corresponding voltage and outputs it to the load, and the energy is directly transmitted through the transformer. According to the specifications, it can be divided into conventional forward, including single-tube forward and double-tube forward. Half-bridge and bridge circuits are both forward circuits.

Forward and flyback circuits have their own characteristics. In the process of designing circuits, in order to achieve the best cost-effectiveness, they can be used flexibly. Generally, flyback can be selected in low-power occasions. A single-tube forward circuit can be used for slightly larger ones, and a double-tube forward circuit or a half-bridge circuit can be used for medium power. A push-pull circuit is used at low voltage, which is the same as the half-bridge working state. For high-power output, a bridge circuit is generally used, and a push-pull circuit can also be used for low voltage. Flyback

power supply is widely used in small and medium-power power supplies because of its simple structure and the elimination of an inductor that is about the same size as the transformer. In some introductions, it is mentioned that the flyback power supply can only achieve a few tens of watts. If the output power exceeds 100 watts, there is no advantage and it is difficult to achieve. I think this is generally true, but it cannot be generalized. The TOP chip of PI company can achieve 300 watts. There are articles that introduce flyback power supplies that can achieve thousands of watts, but I have never seen the real thing. The output power is related to the output voltage.

The leakage inductance of the flyback power supply transformer is a very critical parameter. Since the flyback power supply requires the transformer to store energy, in order to make full use of the transformer core, it is generally necessary to open an air gap in the magnetic circuit. The purpose is to change the slope of the core hysteresis loop so that the transformer can withstand large pulse current impacts without the core entering a saturated nonlinear state. The air gap in the magnetic circuit is in a high magnetic resistance state, and the leakage magnetic field generated in the magnetic circuit is much greater than the completely closed magnetic circuit. The pulse voltage

connection line should be as short as possible, including the input switch tube to the transformer connection line and the output transformer to the rectifier tube connection line. The pulse current loop should be as small as possible, such as the input filter capacitor positive to the transformer to the switch tube return capacitor negative. The output part of the transformer output terminal to the rectifier tube to the output inductor to the output capacitor return transformer circuit should be as close to the input terminal of the switching power supply as possible. The input line should avoid being parallel to other circuits and should be avoided. The Y capacitor should be placed at the ground terminal of the chassis or the FG connection terminal. The common mode inductor should be kept at a certain distance from the transformer to avoid magnetic coupling. Generally,

two output capacitors can be used, one close to the rectifier tube and the other close to the output terminal, which can affect the output ripple index of the power supply. The effect of two small-capacity capacitors in parallel should be better than using a large-capacity capacitor. The heating device should be kept at a certain distance from the electrolytic capacitor to extend the life of the whole machine. The electrolytic capacitor is the life of the switching power supply. For example, the transformer, power tube, and high-power resistor should be kept at a distance from the electrolytic capacitor. Heat dissipation space must also be left between the electrolytic capacitors. If conditions permit, it can be placed at the air inlet.

2. Some principles of printed circuit board wiring

When designing a printed circuit board, the impact of interference on the system should be taken into consideration. The analog and digital parts of the circuit should be strictly separated. The core circuit should be protected. The system ground wire should be surrounded and the wiring should be as thick as possible. The power supply should add a filter circuit, DC-DC isolation should be used, and the signal should use photoelectric isolation. An isolated power supply should be designed. The parts that are prone to interference (such as clock circuits, communication circuits, etc.) and the parts that are easily interfered with (such as analog sampling circuits, etc.) should be analyzed, and measures should be taken for these two types of circuits. Suppression measures should be taken for interfering components, isolation and protection measures should be taken for sensitive components, and they should be spaced and electrically separated. When designing at the board level, it should also be noted that the components should be placed away from the edge of the printed circuit board, which is beneficial for protecting against air discharge. See Figure 1 for the schematic design of the sample circuit.

Figure 1 Schematic diagram

The minimum line spacing is only suitable for signal control circuits and low-voltage circuits with voltages below 63V. When the line voltage is greater than this value, the line spacing can generally be taken according to the empirical value of 500V/1mm.

Method 1: The circuit board slotting method mentioned above is suitable for some occasions where the spacing is not enough. By the way, this method is also often used as a protective discharge gap, which is common in the tail plate of the TV picture tube and the AC input of the power supply. This method has been widely used in module power supplies and can achieve good results under potting conditions.

Method 2: Pad insulation paper, which can be made of insulating materials such as green shell paper, polyester film, and polytetrafluoroethylene oriented film. Generally, general power supplies use green shell paper or polyester film to pad between the circuit board and the metal casing. This material has high mechanical strength and a certain ability to resist moisture. Polytetrafluoroethylene oriented film is widely used in module power supplies due to its high temperature resistance. Insulating film can also be padded between components and surrounding conductors to improve insulation resistance.

The aluminum substrate has the following characteristics due to its own structure: excellent thermal conductivity, single-sided copper bonding, devices can only be placed on the copper bonding surface, and electrical connection holes cannot be opened, so jumpers cannot be placed like single-sided boards.

SMD devices, switch tubes, and output rectifier tubes are generally placed on aluminum substrates to conduct heat away through the substrate. The thermal resistance is very low and high reliability can be achieved. The transformer adopts a flat SMD structure and can also dissipate heat through the substrate. Its temperature rise is lower than the conventional one. The same specification transformer adopts an aluminum substrate structure to obtain a larger output power. Aluminum substrate jumpers can be handled by bridging. Aluminum substrate power supplies are generally composed of two printed circuit boards, and the other board is placed on the control circuit. The two boards are physically connected to form a whole.

Due to the excellent thermal conductivity of aluminum substrate, it is difficult to manually solder a small amount of soldering. Solder cooling too quickly is prone to problems. There is a simple and practical method. Turn over an ordinary electric iron (preferably with a temperature control function) for ironing clothes, ironing side up, fix it, adjust the temperature to about 150℃, put the aluminum substrate on the iron, heat it for a while, and then stick and solder the components according to the conventional method. The iron temperature should be suitable for the device to be easy to solder. If it is too high, the device may be damaged or even the copper foil of the aluminum substrate may peel off. If the temperature is too low, the welding effect is not good. It should be flexibly controlled.

3. Some matters about copper foil routing of printed circuit boards

Routing current density: Most electronic circuits are now made of copper foil bonded to insulating boards. The copper foil thickness of commonly used circuit boards is 35μm. The current density value of the routing can be taken according to the empirical value of 1A/mm. The specific calculation can be referred to the textbook. In principle, the line width should be greater than or equal to 0.3mm to ensure the mechanical strength of the routing. The copper foil thickness is 70μm. Circuit boards are also commonly used in switching power supplies, so the current density can be higher.

Some products in the module power supply line also use multi-layer boards, which are mainly convenient for integrating power devices such as transformer inductors, optimizing wiring, and heat dissipation of power tubes. It has the advantages of good consistency of process and good heat dissipation of transformers, but its disadvantages are high cost and poor flexibility, and it is only suitable for industrial large-scale production.

Single-sided boards, almost all general switching power supplies in the market use single-sided circuit boards, which have the advantage of low cost. Some measures in design and production process can also ensure its performance.

In order to ensure good welding mechanical structure performance, the single-sided board pad should be slightly larger to ensure good bonding between the copper foil and the substrate, so as not to peel off or break off the copper foil when it is vibrated. Generally, the width of the solder ring should be greater than 0.3mm. The pad hole diameter should be slightly larger than the device pin diameter, but not too large to ensure the shortest distance between the pin and the pad by solder connection. The size of the pad hole should not hinder normal inspection. The pad hole diameter is generally 0.1-0.2mm larger than the pin diameter. Multi-pin devices can also be larger to ensure smooth inspection.

Components on a single-sided board should be close to the circuit board. For devices that require overhead heat dissipation, sleeves should be added to the pins between the device and the circuit board, which can play a dual role of supporting the device and increasing insulation. It is necessary to minimize or avoid the impact of external force on the connection between the pad and the pin to enhance the firmness of welding. The heavier components on the circuit board can increase the support connection points to strengthen the connection strength between the circuit board, such as transformers and power device heat sinks.

The double-sided board pad has a higher strength because the hole has been metallized. The solder ring can be smaller than the single-sided board, and the pad hole diameter can be slightly larger than the pin diameter, because it is conducive to the solder solution to penetrate through the solder hole to the top pad during the welding process to increase welding reliability.

4. Processing of high-current routing

The line width can be processed according to the previous post. If the width is not enough, it can generally be solved by tinning the routing to increase the thickness. There are many ways to solve this problem.

1. Set the routing to the pad attribute, so that the routing will not be covered by solder resist when the circuit board is manufactured, and will be tinned during hot air leveling.
2. Place a pad at the wiring location and set the pad to the shape of the required routing. Be careful to set the pad hole to zero.
3. Place the wire on the solder mask layer. This method is the most flexible, but not all circuit board manufacturers will understand your intentions. Textual instructions are required. No solder resist will be applied

to the part where the wire is placed on the solder mask layer.

...

The wiring from one wiring layer to another is generally connected by vias, and it is not suitable to be realized through the device pin pads, because this connection relationship may be destroyed when the device is inserted. In addition, there should be at least 2 vias for every 1A current passing through. The via aperture should be greater than 0.5mm in principle, and generally 0.8mm can ensure processing reliability.

5. Application of aluminum substrates in switching power supplies and multilayer printed circuit boards in switching power supply circuits

Aluminum substrates (metal-based heat sinks (including aluminum substrates, copper substrates, and iron substrates)) are a unique metal-based copper-clad laminate with good thermal conductivity, electrical insulation and mechanical processing properties. Aluminum-based copper-clad laminates are a metal circuit board material, composed of copper foil, thermal conductive insulation layer and metal substrate. Its structure is divided into three layers:

Cireuitl.Layer circuit layer: equivalent to the copper-clad laminate of ordinary PCB, the circuit copper foil thickness is loz to 10oz.
DielcctricLayer insulation layer: The insulation layer is a layer of low thermal resistance thermal conductive insulation material. The thickness is 0.003" to 0.006" inches. It is the core technology of aluminum-based copper-clad laminates and has been UL certified.
The base layer is a metal substrate, generally aluminum or optional copper. Aluminum-based copper-clad laminates and traditional epoxy glass cloth laminates, etc., the mainstream on the market is Foslite aluminum substrate.

The circuit layer (i.e. copper foil) is usually etched to form a printed circuit so that the various components of the assembly are interconnected. In general, the circuit layer is required to have a large current carrying capacity, so a thicker copper foil should be used, with a thickness of generally 35μm~280μm; the thermal conductive insulation layer is the core technology of the aluminum substrate. It is generally composed of a special polymer filled with special ceramics, with low thermal resistance, excellent viscoelastic properties, resistance to thermal aging, and the ability to withstand mechanical and thermal stress. The thermal conductive insulation layer of the high-performance aluminum substrate produced by the company uses this technology, which makes it have extremely excellent thermal conductivity and high-strength electrical insulation performance; the metal base is the supporting component of the aluminum substrate, which is required to have high thermal conductivity. It is generally an aluminum plate, and a copper plate can also be used (where the copper plate can provide better thermal conductivity), which is suitable for conventional mechanical processing such as drilling, punching and cutting.

The aluminum substrate has the following characteristics due to its own structure: very good thermal conductivity, single-sided copper bonding, devices can only be placed on the copper bonding surface, and electrical wiring holes cannot be opened, so jumpers cannot be placed as in a single-sided board.

SMD devices, switch tubes, and output rectifier tubes are generally placed on aluminum substrates. The heat is conducted out through the substrate, and the thermal resistance is very low, which can achieve high reliability. The transformer adopts a flat SMD structure, and can also dissipate heat through the substrate. Its temperature rise is lower than the conventional one. The same specification transformer adopts an aluminum substrate structure to obtain a larger output power. The jumper of the aluminum substrate can be handled by bridging. The aluminum substrate power supply is generally composed of two printed circuit boards, and the other board is placed on the control circuit. The two boards are physically connected to form a whole.

Due to the excellent thermal conductivity of the aluminum substrate, it is difficult to manually solder a small amount of solder. The solder cools too quickly and is prone to problems. There is a simple and practical method. Turn over an ordinary electric iron for ironing clothes (preferably with a temperature control function), iron the surface upward, fix it, adjust the temperature to about 150℃, put the aluminum substrate on the iron, heat it for a while, and then stick and solder the components according to the conventional method. The temperature of the iron is suitable for the device to be easy to solder. If it is too high, the device may be damaged, or even the copper skin of the aluminum substrate may peel off. If the temperature is too low, the welding effect is not good, and it must be flexibly controlled.

In recent years, with the application of multi-layer circuit boards in switching power supply circuits, printed circuit transformers have become possible. Due to the small interlayer spacing of multi-layer boards, the window section of the transformer can be fully utilized. One or two printed coils composed of multi-layer boards can be added to the main circuit board to achieve the purpose of utilizing the window and reducing the line current density. Due to the use of printed coils, manual intervention is reduced, and the transformer has good consistency, flat structure, low leakage inductance, and good coupling. Open core, good heat dissipation conditions. Due to its many advantages, it is conducive to mass production, so it has been widely used. However, the initial investment in research and development is large, and it is not suitable for small-scale production.

Switching power supplies are divided into two forms, isolated and non-isolated. Here we mainly talk about the topological form of isolated switching power supplies. In the following text, unless otherwise specified, all refer to isolated power supplies. Isolated power supplies can be divided into two categories according to different structural forms: forward and flyback. Flyback refers to the working state in which the secondary side is cut off when the primary side of the transformer is turned on, and the transformer stores energy. When the primary side is cut off, the secondary side is turned on, and the energy is released to the load. Generally, conventional flyback power supplies have more single tubes, and double tubes are not common. Forward means that when the primary side of the transformer is turned on, the corresponding voltage is sensed on the secondary side and output to the load, and the energy is directly transferred through the transformer. According to the specifications, it can be divided into conventional forward, including single-tube forward and double-tube forward. Half-bridge and bridge circuits are both forward circuits.

Forward and flyback circuits have their own characteristics. In the process of designing circuits, they can be used flexibly to achieve the best cost performance. Generally, flyback can be selected in low-power occasions. A single-tube forward circuit can be used for slightly larger power, a double-tube forward circuit or a half-bridge circuit can be used for medium power, and a push-pull circuit is used at low voltage, which is the same as the working state of the half-bridge. For high-power output, a bridge circuit is generally used, and a push-pull circuit can also be used for low voltage.

Flyback power supply is widely used in small and medium-power power supplies because of its simple structure and the elimination of an inductor that is about the same size as the transformer. In some introductions, it is mentioned that the power of the flyback power supply can only reach tens of watts, and there is no advantage if the output power exceeds 100 watts, and it is difficult to achieve. I think this is generally the case, but it cannot be generalized. PI's TOP chip can achieve 300 watts. There are articles that introduce flyback power supplies that can achieve thousands of watts, but I have never seen the real thing. The output power is related to the output voltage.

The leakage inductance of the flyback power supply transformer is a very critical parameter. Since the flyback power supply requires the transformer to store energy, in order to make full use of the transformer core, it is generally necessary to open an air gap in the magnetic circuit. The purpose is to change the slope of the core hysteresis loop so that the transformer can withstand large pulse current impacts without causing the core to enter a saturated nonlinear state. The air gap in the magnetic circuit is in a high magnetic resistance state, and the leakage magnetic flux generated in the magnetic circuit is much greater than the completely closed magnetic circuit.

The coupling between the primary poles of the transformer is also a key factor in determining the leakage inductance. The primary pole coils should be as close as possible. The sandwich winding method can be used, but this will increase the distributed capacitance of the transformer. When selecting the core, try to use a core with a relatively long window to reduce the leakage inductance. For example, the effect of using EE, EF, EER, and PQ cores is better than that of EI.

Regarding the duty cycle of the flyback power supply, in principle, the maximum duty cycle of the military power supply should be less than 0.5, otherwise the loop will not be easily compensated and may be unstable, but there are some exceptions, such as the TOP series chips launched by the American PI company, which can work under the condition of a duty cycle greater than 0.5. The duty cycle is determined by the ratio of the primary and secondary turns of the transformer. My opinion on flyback is to first determine the reflected voltage (the voltage value reflected by the output voltage to the primary side through the transformer coupling). Within a certain voltage range, the reflected voltage increases, the working duty cycle increases, and the switch tube loss decreases.

Next, let's talk about the duty cycle of the flyback power supply (I am concerned about the reflected voltage, which is consistent with the duty cycle). The duty cycle is also related to the withstand voltage of the switch tube. Some early flyback power supplies used relatively low withstand voltage switch tubes, such as 600V or 650V as the switch tube for the AC 220V input power supply. Perhaps it was related to the production process at the time. High withstand voltage tubes were not easy to manufacture, or low withstand voltage tubes had more reasonable conduction losses and switching characteristics. For example, the reflected voltage of this line cannot be too high, otherwise, in order to make the switch tube work within a safe range, the power absorbed by the circuit loss is also considerable. Now, due to the improvement of the manufacturing process level of MOS tubes, general flyback power supplies use 700V or 750V or even 800-900V switch tubes. These two types have their own advantages and disadvantages:

The first type: Disadvantages: weak overvoltage resistance, small duty cycle, and large transformer primary pulse current. Advantages: small transformer leakage inductance, low electromagnetic radiation, high ripple index, small switch tube loss, and conversion efficiency is not necessarily lower than the second type.

Category 2: Disadvantages: The switch tube has larger loss, the transformer leakage inductance is larger, and the ripple is worse. Advantages: Stronger overvoltage resistance, large duty cycle, lower transformer loss, and higher efficiency.

6. There is another determining factor for the reflected voltage of the flyback power supply.

The reflected voltage of the military switching power supply is also related to a parameter, that is, the output voltage. The lower the output voltage, the larger the transformer turns ratio, the larger the transformer leakage inductance, and the higher the voltage the switch tube withstands. It is possible to break down the switch tube and the absorption circuit consumes more power, which may cause the absorption circuit power device to fail permanently. Care must be taken in the optimization process of designing a low-voltage output low-power flyback power supply. There are several ways to deal with it:

1. Use a magnetic core with a larger power level to reduce leakage inductance, which can improve the conversion efficiency of the low-voltage flyback power supply, reduce losses, reduce output ripple, and improve the cross-regulation rate of multi-channel output power supplies. It is generally common in switching power supplies for household appliances, such as optical disc players, DVB set-top boxes, etc.

2. If conditions do not allow the magnetic core to be increased, the reflected voltage can only be reduced and the duty cycle can only be reduced. Reducing the reflected voltage can reduce leakage inductance but may reduce power conversion efficiency. These two are contradictory. A replacement process is necessary to find a suitable point. During the transformer replacement experiment, the reverse peak voltage of the primary side of the transformer can be detected to reduce the width and amplitude of the reverse peak voltage pulse as much as possible, which can increase the working safety margin of the converter. Generally, the reflected voltage is more suitable at 110V.

3. Enhance coupling, reduce loss, adopt new technology and winding process. To meet safety regulations, the transformer will take insulation measures between the primary and secondary sides, such as padding with insulating tape and adding insulating end empty tape. These will affect the leakage inductance performance of the transformer. In actual production, the winding method of the primary winding wrapped around the secondary can be adopted. Or the secondary is wound with triple insulation wire, and the insulation between the primary and secondary is eliminated, which can enhance coupling, and even wide copper foil can be used for winding.

The magnetic core of the flyback power transformer is working in a unidirectional magnetization state, so the magnetic circuit needs to have an air gap, similar to a pulsating DC inductor. Part of the magnetic circuit is coupled through an air gap. I understand the principle of opening the air gap as follows: Since the power ferrite also has a working characteristic curve (hysteresis loop) that is similar to a rectangle, the Y axis on the working characteristic curve represents the magnetic induction intensity (B). The current production process generally has a saturation point above 400mT. Generally, this value should be taken in the design at 200-300mT. The X axis represents the magnetic field intensity (H). This value is proportional to the magnetizing current intensity. Opening an air gap in the magnetic circuit is equivalent to tilting the magnetic hysteresis loop of the magnet toward the X axis. Under the same magnetic induction intensity, it can withstand a larger magnetizing current, which is equivalent to storing more energy in the magnetic core. This energy is discharged to the load circuit through the secondary of the transformer when the switch tube is turned off. The air gap in the magnetic core of the flyback power supply has two functions.

The transformer of the flyback power supply works in a unidirectional magnetization state. It not only transfers energy through magnetic coupling, but also bears the multiple functions of voltage conversion input and output isolation. Therefore, the air gap needs to be handled very carefully. If the air gap is too large, the leakage inductance will increase, the hysteresis loss will increase, and the iron loss and copper loss will increase, affecting the overall performance of the power supply. If the air gap is too small, the transformer core may be saturated, resulting in power supply damage.

The so-called continuous and intermittent modes of the flyback power supply refer to the working state of the transformer. Under full load, the transformer works in a working mode of complete or incomplete energy transfer. Generally, it should be designed according to the working environment. Conventional flyback power supplies should work in continuous mode, so that the losses of the switch tube and the line are relatively small, and the working stress of the input and output capacitors can be reduced, but there are some exceptions. Due to the characteristics of the manufacturing process, high reverse voltage diodes have a long reverse recovery time and low speed. In the continuous current state, the diode recovers when there is a forward bias. The energy loss during reverse recovery is very large, which is not conducive to improving the performance of the converter. At the least, the conversion efficiency is reduced, the rectifier tube is seriously heated, and at the worst, the rectifier tube is even burned. Since the diode is reverse biased under zero bias in the intermittent mode, the loss can be reduced to a relatively low level. The

flyback switching power supply transformer should work in the continuous mode, which requires a relatively large winding inductance. Of course, there is a certain degree of continuity. It is unrealistic to pursue absolute continuity excessively. It may require a large magnetic core and a lot of coil turns, accompanied by large leakage inductance and distributed capacitance, which may not be worth the loss. So how to determine this parameter? Through many practices and analysis of the designs of peers, I believe that when the nominal voltage is input, the output reaches 50%~60% and the transformer transitions from intermittent to continuous state is more appropriate.