Thermal Design of Switching Power Supplies

If the temperature rise inside the switching power supply is too high, it will cause the failure of temperature-sensitive semiconductor devices, electrolytic capacitors and other components. When the temperature exceeds a certain value, the failure rate increases exponentially. Statistics show that the reliability of electronic components will decrease by 10% for every 2°C increase in temperature; the life span at a temperature rise of 50°C is only 1/6 of that at a temperature rise of 25°C. In addition to electrical stress, temperature is the most important factor affecting the reliability of the switching power supply. High-frequency switching power supplies have high-power heating devices, and temperature is one of the most important factors affecting their reliability. A complete thermal design includes two aspects: one is how to control the heat generated by the heat source; the other is how to dissipate the heat generated by the heat source so that the temperature rise of the switching power supply is controlled within the allowable range to ensure the reliability of the switching power supply. The following is an explanation from two aspects.

(1) Design to control heat generation

The main heat generating components in the switching power supply are semiconductor switch tubes, power diodes, high-frequency transformers, filter inductors, etc. Different components have different ways to control heat generation. The power tube is one of the devices with the largest heat generation in the high-frequency switching power supply. Reducing its heat generation can not only improve the reliability of the power tube, but also improve the reliability of the switching power supply and the mean time between failures (MTBF). The heat generation of the switch tube is caused by loss. The loss of the switch tube consists of two parts: switching process loss and conduction loss. Reducing the conduction loss can be achieved by selecting a switch tube with low conduction resistance. The switching process loss is caused by the gate charge size and switching time. Reducing the switching process loss can be achieved by selecting a device with faster switching speed and shorter recovery time. But more importantly, it is necessary to reduce the loss by designing a better control method and buffer technology. For example, the use of soft switching technology can greatly reduce this loss. To reduce the heat generation of the power diode, there is generally no better control technology to reduce the loss for AC rectifiers and buffer diodes. The loss can be reduced by selecting high-quality diodes. For the rectification of the secondary side of the transformer, a more efficient synchronous rectification technology can be selected to reduce the loss. For the loss caused by high-frequency magnetic materials, the skin effect should be avoided as much as possible. The influence caused by the skin effect can be solved by winding multiple strands of fine enameled wire.

(2) Heat dissipation design of switching power supply

When the power MOSFET is turned on, there is a certain voltage drop, which means that the device has a certain loss, which will cause the temperature of the chip to rise. However, the heating of the device is related to its heat resistance and heat dissipation conditions. Therefore, the power consumption of the device has a certain tolerance. Its value can be expressed as:

Where Tj——rated junction temperature (Tj＝150℃);

Tc——shell temperature:

RT - Steady-state thermal resistance from junction to case.

Tj represents the heat resistance of the device, Tc and RT represent the heat dissipation conditions of the device, and PD is the heat generation of the device. It must strike a balance between the heat resistance of the device and the heat dissipation conditions.

There are three basic ways to dissipate heat: heat conduction, heat radiation, and heat convection. Depending on the way of dissipating heat, you can choose natural heat dissipation; force a heat sink; or choose forced air cooling: add a fan. The installation of a heat sink mainly utilizes heat conduction and heat convection, that is, all heat-generating components are fixed on the heat sink, and the heat is transferred to the heat sink by conduction. The heat on the heat sink is then carried out of the chassis by air through energy flow heat exchange. The actual heat dissipation situation is a combination of the three heat transfer methods, which can be expressed uniformly by Newton's formula:

Where S is the heat dissipation surface area;

K——Surface heat dissipation coefficient.

The surface heat dissipation coefficient is usually determined by experiments, and data can be found in general engineering fluid mechanics. It unifies all three forms of heat transfer.

Through Φ=KST, after calculating the dissipated power, the heat dissipation surface area S can be determined according to the allowable temperature rise r, and the heat sink to be selected can be determined accordingly. This calculation is of great significance for improving the reliability, power density, performance-price ratio, etc. of the switching power supply. If forced air cooling is adopted, such as installing a fan, then for the rectifier module, the MTBF of the fan is the lowest among all components, and it has always been the bottleneck restricting the improvement of the MTBF of the rectifier module. Therefore, it is of great significance to take various measures to improve the heat dissipation efficiency to extend the life of the fan.