Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit

In recent years, in order to further improve global energy conservation, global regulatory agencies have proposed new efficiency standards. With a series of new requirements issued by the U.S. Department of Energy (DOE), manufacturers must improve the efficiency of existing independent power supply products to meet DOE VI level standards before they can be sold in the U.S. market. In addition, manufacturers also need to design products that meet other energy specifications, such as the EU CoC V5Tier2 specifications.

Authors: Zhihong Yu, Walter Yeh

In recent years, in order to further improve global energy conservation, global regulatory agencies have proposed new efficiency standards. With a series of new requirements issued by the U.S. Department of Energy (DOE), manufacturers must improve the efficiency of existing independent power supply products to meet DOE VI level standards before they can be sold in the U.S. market. In addition, manufacturers also need to design products that meet other energy specifications, such as the EU CoC V5Tier2 specifications.

In order to improve the efficiency of the AC-to-DC adapter, the output freewheeling Schottky diode is replaced with a MOSFET-based synchronous rectifier controller (SR), which can usually increase the efficiency by 2 to 3% or higher. It has also been found that the use of SR helps to save the cost of the diode heat sink and the cost of manual assembly. Designers can also use cheaper primary MOSFETs or thinner output cables to save costs and still achieve the target efficiency.

Due to space limitations, this article cannot cover all the details of SR design. Instead, a few practical topics are selected for discussion when engineers design synchronous rectifier circuits.

Continuous conduction mode (CCM) of SR

In Figure 1, the flyback SR controller is used to drive the secondary MOSFET switch in the AC/DC adapter. Here, the flyback controller can operate in critical conduction mode (CrM), continuous conduction mode (CCM) or discontinuous conduction mode (DCM).

Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit
Figure 1: A typical block diagram of a flyback power supply used in a fast charger

The adapter runs in CCM mode when it is started or fully loaded. When the main switch is trying to turn on, the current in the SR switch is set and cannot drop to zero. Therefore, it is necessary to prevent the breakdown of the primary side to the secondary side from causing high-voltage spikes and potential damage, and therefore the SR needs to be turned off quickly. The solution of MPS is to adjust the VG voltage of the SR switch to keep the VDS of the MOSFET constant. As the current decreases during the CCM mode, the VG voltage of the driver also decreases until the MOSFET runs in the linear operating region (see Figure 3). Therefore, when the voltage is finally reversed, the driver will quickly turn off based on a very low VG voltage to ensure safe operation in CCM mode. Because it is not affected by the input conditions of the line, this is a stable control method. In addition, by maximizing the on-time of the SR MOSFET and minimizing the on-time of the body diode, the best efficiency can be ensured. The SR controller of MPS can not only support CCM mode, but also DCM and CrM mode.

Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit
Figure 2: Primary and secondary current waveforms in CCM mode

For a detailed description of SR design and operation in CCM compatibility mode of MPS, please refer to the AN077 application note. 1.

Influence of MOSFET package inductance in CCM mode and CrM mode

When the secondary current switches, there will always be some switch rise/fall time (as shown in Figure 2), which is determined by the input/output, transformer turns ratio and inductance. The MOSFET package inductance also affects the turn-off of the secondary current.

As the secondary current starts to change polarity and turn off (t1 in Figure 4), the MOSFET package inductance (Ls) will generate an instantaneous voltage on the detected Vds, as shown in equations (1) and (2):

Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit

Among them, dc is the average DC input, n is the transformer turns ratio, and Ls is the leakage inductance.

Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit
Figure 3: MPS SR controller operating principle

For the TO220 packaged MOSFET, the package inductance can be as high as 6.4nH at a frequency of 100kHz, and the Vlk can be as high as several hundred mV, reaching the turn-off threshold of the SR controller, making the SR controller turn off the gate (starting from t1). Since the t1 turn-off time is relatively early, a slightly higher package inductance helps prevent breakdown, especially under deep CCM conditions.

For various circuit designs, we may see different turn-off waveforms in CCM mode (see Figure 4a and Figure 4b). As shown in Figure 4a, the current drops to zero, but the SR is not completely turned off. Therefore, cross conduction may occur and will be reflected in the reverse current. The relatively best design is that SR can be turned off before the secondary current becomes zero (t2), as shown in Figure 4b. More noteworthy is that, as shown in Figure 4c, in the CrM mode, when the secondary side current is almost zero, the SR controller will turn off accordingly, which means that there is always a reverse current dI / dt * Toff .

When the package inductance of the MOSFET is very small (such as QFN or SOIC package), the SR gate will be more delayed relative to turn off. Even if Vg is reduced under the control of Vds regulation, the reverse current is still greater than that of a MOSFET with a higher package inductance. This has nothing to do with the Vds control introduced in topic 1.

● Choose SR MOSFET with very low Qg (to speed up turn-off).

Add an RC snubber absorbing circuit to the SR MOSFET (to absorb reverse voltage spikes). Use SR controller with high shutdown current. Increasing transformer leakage inductance slows down the secondary current dI / dt during turn-off (but will cause higher primary MOSFET voltage spikes) to slow down the rising slope (loss of efficiency) when the primary MOSFET turns on. Use SR controller with higher Vds control voltage (MP6902 using MPS in Figure 2 is 70mV). In the case of a higher Vds control voltage, the MOSFET can enter a deeper linear region, and the Vg will reach a very low level before the switch is turned off, thereby quickly turning off.

Ringing-advantages and disadvantages

When the MOSFET is turned on and off, the discrete inductance generated in the PCB layout and the system and the parasitic capacitance in the components will cause some ringing. If you can’t adapt to the impact of ringing, it may reduce efficiency at the slightest level, and cause some fatal problems at the worst.

The problem caused by the ringing is shown in Figure 4. When the secondary current drops to zero, the primary switching voltage Vds will resonate between the main inductance of the transformer and the MOSFET Cds, and this resonant voltage will be refracted to the secondary side. Usually, this resonance valley should not touch the ground plane, but sometimes the resonance valley may drop to the turn-on threshold of SR. This may be due to factors such as the reverse recovery of the diode in the primary RCD snubber.

Since the slope of the Vds voltage resonance is always much lower than the actual switch-off slope (due to the larger inductance of the main inductance), MPS’s MP6908 uses a unique adjustable slope pin to help determine when the secondary side MOS Real turn-off, and when is the normal Vds voltage resonance (as shown in Figure 4).

Figure 4: SR waveform of potential false opening during degaussing ringing

Replace Schottky diodes according to actual needs

Although the advantages of SR have been widely accepted, changing the design of Schottky diodes to a design scheme using SR drivers and MOSFETs still needs to add many components to the BOM and requires re-certification.

Another solution is to integrate the SR MOSFET into the SR driver IC to create a compact package to replace the Schottky diode without any changes to the transformer. This new design minimizes the BOM change (see Figure 5). This solution is called an ideal diode solution.

The advantages of MPS’s new ideal diode are as follows:

● Minimal BOM and circuit board space.

● The Schottky diode can be directly replaced on the high side or the low side without auxiliary winding.

● Optimized integrated gate driver.

● Optimize MOSFETs for different power levels and rated voltages.

● Flexible SMT and through-hole packaging options.

Why is MPS MP6908 the choice for actual SR control design?

MP6908 is MPS’s latest SR control IC, and there will be a series of ideal diode solutions based on MP6908 controller in the future. Some of the main functions of this controller IC include:

● No auxiliary winding for high-side or low-side rectification is required.

● Support DCM, quasi-resonant and CCM operation modes.

It supports a wide output range as low as 0V (even when the output is short-circuited, the SR keeps supplying power, and the short-circuit current does not flow through the body diode of the MOSFET). Ring detection can prevent false turn-on. Ultra-high speed 15ns propagation delay and 30ns turn-off delay.

Analysis of continuous conduction mode of SR when designing synchronous rectifier circuit
Figure 5: MP6908 controller and ideal diode application circuit for low-side and high-side

Summarize

This article introduces the synchronous rectifier (SR) design related to the actual engineering situation. By learning more about terminal applications, MPS can define and create better SR control ICs.

The Links:   TD210N12 G190ETN01-204

Related Posts

Leave a Reply

Your email address will not be published. Required fields are marked *