[Introduction]The electrical energy provided by the grid is AC, but most of the equipment we use requires DC, which means that the AC/DC power source that does this conversion is one of the most common loads on the energy grid. As the world focuses on energy efficiency to protect the environment and manage operating costs, efficient operation of these power sources is increasingly important.

Efficiency is easy to understand, measured as the ratio between the input power and the power supplied to the load. However, the input power factor also has a large effect. Power factor (PF) describes the ratio between the useful (real) power and the total (apparent) power (kVA) consumed by any AC device (including power supplies). PF measures the effectiveness of converting consumed electrical energy into useful work output.

If the load is purely resistive, PF will be equal to 1, but any reactive components within the load will reduce PF, making the apparent power greater than the useful power, resulting in lower efficiency.

PF less than 1 is caused by the voltage and current being out of phase – this is common with inductive loads. It can also be due to high harmonic content or distortion of the current waveform, which is common in Switch Mode Power Supplies (SMPS) or other types of discontinuous Electronic loads.

Corrected PF

Many SMPSs without PF correction consume higher current than corrected SMPSs, so at power levels above 70W, legislation requires designers to add circuitry to correct PF to a value close to 1. The most common active PF correction (PFC) technique uses a boost converter to convert the rectified power supply to a DC high level, and then uses pulse width modulation (PWM) to regulate the DC level.

While this technique works well and is easy to implement, there are some challenges. Modern efficiency standards (such as the stringent “80+ Titanium Standard”) require high efficiency over a wide power range, with a peak efficiency of 96% at 50% load. Since the DC-DC converter after the PFC stage typically has 2% losses, the linear rectification and the PFC stage itself can only lose 2% – a challenge for diodes in bridge rectifiers.

Figure 1: Traditional (left) and (right) bridgeless totem PFC circuits

However, if the boost diode (D5) is replaced with a synchronous rectifier, the efficiency improves. In addition, only two linear rectifier diodes are required, and synchronous rectifiers (Q3 and Q4) can also be used to further improve efficiency. Known as totem pole PFC (TPPFC), this technology can approach 100% efficiency with ideal inductors and excellent switching. Modern MOSFETs have excellent performance, but are not yet at the level of ideal switching – even when used in parallel. Therefore, wide bandgap semiconductor switches will go hand in hand with totem pole PFC topologies.

coping with loss

With the ever-increasing trend of switching frequencies, dynamic losses in switching devices will have a greater impact. These losses are due to reverse recovery when the MOSFET is configured as a totem-pole high-speed switching leg. When its body diode conducts during the switch “dead time” time, the associated stored charge must be depleted. The losses also come from the switch. charging and discharging of the output capacitor.

In fact, the dynamic losses of silicon-based MOSFETs can be quite large, so designers are increasingly specifying wide-bandgap semiconductor materials such as silicon carbide (SiC) and gallium nitride (GaN) devices in TPPFC applications. Additional benefits of these devices are higher operating frequencies and high temperature capability—two features that are very useful in power supply applications.

Critical conduction mode (CrM) is often the preferred conduction mode for TPPFCs, especially at powers up to a few hundred watts, providing a good compromise between efficiency and EMI performance. Continuous Conduction Mode (CCM) further reduces RMS current and conduction losses in the switch, making the TPPFC suitable for kilowatt-rated power applications.

Even with CrM, the efficiency drops significantly under light load conditions (up to 10%), which presents a real challenge when trying to meet standby or no-load power consumption constraints. One solution is to clamp or “fold back” the maximum allowed frequency, thereby forcing the circuit into DCM under light load conditions, where the higher peak currents are still lower than those in an equivalent CrM implementation.

Combining TPPFC with CrM operation and frequency clamping provides a good mid-power solution with excellent efficiency over the entire load range, especially when the WBG switch is used for high frequency bridge legs.

other challenges

Once the efficiency challenge has been addressed, there is one final hurdle to overcome. Four active devices need to be driven synchronously, and zero current crossing of the Inductor must be detected to force CrM. The circuit must be able to automatically switch in and out of DCM when needed, and do all of this while maintaining a high power factor and generating a PWM signal to regulate the output. In addition to this, circuit protection (eg overcurrent and overvoltage) is required.

In general, given the complexity involved, the best approach is to implement the control algorithm in a microcontroller. However, this approach can be expensive and requires generating and debugging code, an area that many designers want to avoid.

No-code solution for TPPFC with CrM

A fully integrated TPPFC control solution offers many advantages, including the ability to increase performance levels, reduce design time, and reduce design risk while eliminating the need for a microcontroller and associated code.

One such integrated solution is the mixed-signal TPPFC Controller NCP1680 from onsemi, which operates at constant on-time CrM, ensuring high efficiency over the entire load range. The NCP1680 provides “valley switching” during frequency foldback at light loads, improving efficiency by switching at the lowest voltage. The digital voltage control loop is internally compensated to optimize performance over the entire load range while keeping the design process simple.

Figure 2: The NCP1680 provides a simple and sophisticated no-code TPPFC solution

Housed in a small SOIC-16 package, this innovative controller utilizes a proprietary low-loss method for current sensing and cycle-by-cycle current limiting, providing an excellent level of protection without the need for external Hall-effect sensors, reducing complexity, size, and cost.

All necessary control algorithms are embedded in the IC, providing designers with a low-risk, tried-and-tested solution that delivers high performance at an economical price point.