Tweet

Posted on 01 September 2019

The SmartRectifier in Practical Design

Free Bodo's Power Magazines!

 

  

 

A step forward for secondary-side rectification

Power supply architecture has become a major issue in product design over recent years. At the core of this trend has been the decrease in digital IC voltages, while overall power densities have moved in the opposite direction.

By Mario Battello, International Rectifier

 

Change has also been driven by the constantly developing nature of end products, all with different current and voltage requirements, and many representing not just evolutions, but whole new device categories. LCD monitors, laptop computers, home-theatre setups, “headless” mini-PCs, games consoles, LCD and PDP TVs; all have unique requirements that the power designer needs to understand.

Last but not least has been a growing awareness of the need to conserve energy so as to make prudent use of natural resources. Not far down the road, legislation, such as Europe’s EuP (energy-using products) Directive, will lay down the law on requirements for efficient use of energy. These will sit alongside specifications such as EnergyStar and the IEA’s 1W Standby initiative.

As a consequence, power-supply-rectification architectures have, like the products in which they are used, also come a long way. The basic half-wave diode rectifier, with its bulky heatsinks and high power dissipation, has given way to the use of more efficient synchronous- rectification schemes using MOSFETs. A typical cost-effective flyback converter serves as an example. In Figure 1, the schematic on the left utilizes a diode rectifier whereas the schematic on the right uses a synchronous-rectifier MOSFET. Assuming that the voltage across the diode in conduction mode is 0.6V, that the FET RDSon @ 100 C is 10mOhm, and that the IRECTRMS is 5A the power loss of each device is as appears in Table 1.

Synchronous-rectifier MOSFET replaces a diode rectifier

Under the same operating conditions, the synchronous-rectifier MOSFET has a smaller power dissipation compared with the diode rectifier during the on phase. Due to lower device temperature the size of the solution can be reduced so increasing power density. For larger currents, using the synchronous-rectifier MOSFET to perform secondary side rectification is necessary.

Comparing the characteristics of a diode rectifier and synchronous rectifier MOSFET

A subsidiary benefit is that the components can be smaller, improving power density. And the advantages become accentuated at higher current levels, since the diode on-voltage is fixed.

The challenge in implementing such FET architectures, however, is that they require relatively complex circuitry to control the MOSFET switch. One approach is to derive the required timing signals from the primary-side waveform. However, practical circuits that use this principle tend to perform poorly under light load conditions, and struggle to comply with targets such as 1W Standby.

The alternative is to use the secondary side as the source of control. Flyback converters commonly include a current transformer on the secondary side for this purpose (see Figure 2).

Conventional current sensing architecture

This technique improves overall system efficiency, but still has its drawbacks. The topology triggers the MOSFET’s off interval after the reverse current has developed to a sufficient amplitude - a phenomenon analogous to a diode with a stately recovery time. The resultant circulating current does nothing but reduce the overall supply efficiency, which adds to the design’s thermal load, and increases the output ripple for a given charge storage capacity.

As a result of these shortcomings, International Rectifier has developed a topology known as SmartRectifier, which works by measuring the voltage across the synchronous rectifier switch. The result is an efficient, fast design, with low parts count.

The principle is to ensure that the synchronous rectifier FET is switched very near the current’s zero crossing. IR has introduced an IC implementation of the architecture, the IR1167, which includes a pair of high-speed 200V comparators that sense the FET’s drain-source voltage: during the ON interval, this provides a reflection of the current through the FET’s on-resistance.

Flyback Converter using IR1167 SmartRectifier Control

When used in a typical flyback converter design (see Figure 3), the IC differentially senses this drain to source voltage and compares it to three voltage thresholds – VTH1, VTH2, and VTH3 – to determine the correct time to turn on or turn off the SR switch (see Figure 4).

IR1167 voltage thresholds

The operation of the circuit can be analysed in three stages, starting with the decision to turn the SR switch on. This process begins when the primary switch turns off, forcing current through the parasitic diode in the SR switch. This creates a larger (negative) voltage value than the voltage drop caused by current flowing through the MOSFET on-resistance. When this voltage reaches the VTH2 of the IR1167, the switch will turn on, allowing current to flow from drain to source and decreasing VDS.

For a flyback converter in DCM/CrCM (discontinuous or critical conduction mode, see Figure 5), the rectified current decreases after the switch turns on. The absolute value of VDS also decreases until it reaches VTH1: at this point the switch is turned off again.

Flyback Converter Secondary side DCM-CrCM operation

There are several fine-tuning points within this cycle. First, circuit resonances at switch-on can allow the drain-source voltage to dip below VTH1 and turn the switch off. The IR1167 therefore includes a programmable minimum on-time (MOT) to prevent such spurious effects, simultaneously limiting the minimum duty cycle of the secondary side as well as the maximum duty cycle of the primary side switch.

Second is the danger of a false trigger for the turn-on threshold. This can occur because once the drain-source voltage falls below VTH1, residual current flows through the body diode, possibly allowing the drain-source voltage to reach VTH2. To prevent this, the IR1167 inserts a blanking time after turn-off, during which the IC is inactive. When VDS reaches VTH3 after the blanking period ends, the device resumes operation.

The situation is slightly different in continuous current mode (CCM, see Figure 6). During the conduction phase the current decays and the drain-source voltage decreases. When the primary switch turns back on, the current through the secondary FET rapidly decreases. This forces the drain-source voltage beyond VTH1, turning off the FET, and sending residual current freewheeling back into the primary side.

Flyback Converter Synchronous Rectification turn off waveforms in CCM

This means that timing the turn-off is critical in preventing cross-conduction. Timing the turn-off to match as closely as possible to the point where the current reaches zero also helps to reduce power losses in the switch.

The IR1167 is designed to help designers working under a broad range of circuit design constraints. Its gate drive can source peak currents of 2A and sink up to 7A, into and out of a 10nF gate capacitance. Two versions of the device are available, giving the choice of 10.7V and 14.5V internal clamping voltages.

The IC provides other features required by designers of modern power controllers. These include a sleep mode that reduces standby current to 200µA for compliance with the IEA’s 1W-standby initiative, Energy Star, and CECP.

The SmartRectifier principle produces a number of circuit design benefits for the engineer. Firstly, switching losses are reduced, because the circuit allows current to flow through the body diode before the switch is turned on. The effect is to reduce the gate-charge required for switch-on.

The second consequence – produced by a combination of the architecture and the level of integration of the IC itself – is that the parts count for a typical secondary circuit is reduced to six, including the SmartRectifier controller. This 75% reduction relative to a currenttransformer based design reduces PCB area, BOM and assembly costs: just as importantly for modern consumer goods, it allows designers to reach new levels of power density.

A 1% increase in efficiency also contributes a further power density improvement. Although this may seem like a very small step forward, it should be remembered that the efficiency of such designs has in recent times increased by, on average, around 0.5% per year. A 1% advance therefore represents two years’ development.

Market and regulatory developments mean that power supply design is now well and truly on the map. New developments such as SmartRectifier will allow designers to continue producing innovative products, whilst satisfying increasingly stringent legislation and customer demands in terms of power consumption.

 

 

VN:F [1.9.17_1161]
Rating: 0.0/6 (0 votes cast)

This post was written by:

- who has written 791 posts on PowerGuru - Power Electronics Information Portal.


Contact the author

Leave a Response

You must be logged in to post a comment.