Synchronous Rectification Technology in High-Performance Power Converter Design

As power converters become more common, electronic equipment manufacturers are demanding new features and functions from their power systems, such as lower input and output voltages, higher currents, and faster transient responses.

To meet these demands, in the late 1990s, switching power supply designers began to adopt synchronous rectification (SR) technology—using MOSFETs to replace the rectification function commonly implemented by diodes. SR improves efficiency, thermal performance, power density, manufacturability, and reliability, and can reduce the power system cost of the entire system. This article will introduce the advantages of SR and discuss the challenges encountered in its implementation.

Disadvantages of Diode Rectification

Figure 1 is a schematic diagram of a nonsynchronous and synchronous buck converter. A nonsynchronous buck converter uses a FET and a Schottky diode as the switching device (Figure 1a). When the FET is turned on, energy is transferred to the output inductor and the load. When the FET is turned off, the current in the inductor flows through the Schottky diode. If the load current is higher than half of the ripple current of the output inductor, the converter operates in continuous conduction mode. The Schottky diode is selected based on its forward voltage drop and reverse leakage current characteristics. However, when the output voltage is reduced, the forward voltage of the diode becomes important, which will reduce the efficiency of the converter. Physical characteristics make it difficult to reduce the forward voltage drop of the diode below 0.3V. In contrast, the on-resistance R _DS(ON) of the MOSFET can be reduced by increasing the size of the silicon die or connecting discrete devices in parallel . Therefore, for a given current, using a MOSFET instead of a diode can achieve a much smaller voltage drop than a diode.

This makes SR very attractive, especially in applications that are sensitive to efficiency, converter size, and thermal performance, such as portable or handheld devices. MOSFET manufacturers continue to introduce new MOSFET technologies with lower R _DS(ON) and total gate charge (QG), which make it easier to implement SR in power converter designs.

What is synchronous rectification?

For example, in a synchronous buck converter, efficiency can be improved by replacing the Schottky diode with two low-side MOSFETs (Figure 1b). The two MOSFETs must be driven in complementary mode with a small dead time between their conduction intervals to avoid simultaneous conduction. The synchronous FET operates in the third quadrant because the current flows from the source to the drain. Compared to its nonsynchronous counterpart, the synchronous buck converter always operates in continuous conduction, even at no load.

During the dead time, the inductor current flows through the body diode of the low-side FET. This body diode usually has a very slow reverse recovery characteristic, which reduces the efficiency of the converter. A Schottky diode can be placed in parallel with the low-side FET to bypass the body diode and prevent it from affecting the performance of the converter. The added Schottky diode can have a much lower current rating than the diode in the nonsynchronous buck converter because it only conducts during the short dead time (usually less than a few percent of the switching period) when both FETs are off.

Benefits of Synchronous Rectification

The benefits of using SR in high performance, high power converters are higher efficiency, lower power dissipation, better thermal performance, and the inherent ideal current sharing characteristics when synchronous FETs are connected in parallel, and improved manufacturing yields despite the use of automated assembly processes (higher reliability). As mentioned above, several MOSFETs can be connected in parallel to handle higher output currents. Because the effective R _DS(ON)

in this case is inversely proportional to the number of devices connected in parallel, conduction losses are reduced. Also, R _DS(ON) has a positive temperature coefficient, so the FETs will share the current equally, helping to optimize the heat distribution between the SR devices, which will improve the ability of the device and PCB to dissipate heat, directly improving the thermal performance of the design. Other potential benefits of SR include smaller form factor, open frame structure, higher ambient operating temperature, and higher power density.

Design Tradeoffs in Synchronous Rectification Converters

In low voltage applications, designers often increase the switching frequency to reduce the size of the output inductor and capacitor, thereby minimizing the converter size and reducing the output ripple voltage. This increase in frequency also increases gate drive and switching losses if multiple FETs are connected in parallel, so design tradeoffs must be made based on the specific application. For example, in a high input voltage, low output voltage synchronous buck converter, because the operating condition is that the high-side FET has a lower RMS current than the low-side FET, the high-side FET should be selected with a device with low QG and high R _DS(ON) . For this device, reducing switching losses is more important than conduction losses. In contrast, the low-side FET carries a larger RMS current, so R _DS(ON) should be as low as possible.

Selecting a controller with stronger drive capability in a synchronous converter will reduce switching losses by minimizing the time the FETs spend switching on and off. However, faster rise and fall times can generate high-frequency noise that can cause system noise and EMI issues.

Synchronous Rectification Converter Drive in Isolated

Topology Power converters using isolated topology are used in systems that require isolation between system grounds. Such systems include distributed bus architectures, Power over Ethernet systems, and wireless base stations (Figure 2).

Using SRs in isolated converters can greatly improve their performance. All isolated topologies, including forward, flyback, push-pull, half-bridge, and full-bridge (current and voltage feedback) can perform synchronous rectification. However, providing adequate and timely gate drive signals from the SRs in each topology presents its own challenges.

There are basically two driving schemes for the secondary FETs of isolated topologies: self-driven gate signals derived directly from the secondary transformer windings, and controlled drive gate signals derived from a PWM controller or some other primary reference signal. For a given application, these drives can be implemented in several different ways. Designers should choose the simplest solution that meets the performance requirements.

The self-driven scheme is the simplest and most direct SR drive scheme (Figure 3) and is suitable for topologies where the transformer voltage is not zero for any period of time. Two SR FETs replace the output rectifier diodes, and the voltage generated by the secondary winding drives the gates of the SRs. In most cases, the same topology can achieve higher or lower output voltages by using different transformer turns ratios (NP: NS1: NS2) and properly selecting SR FETs.